M&A in Japan

Erim - 08 omslag Schaik :Erim omslag Schaik 15-09-2008 12:23 Pagina 1 B&T28304 - Erim Omslag van Schaik

M&A IN JAPAN 141 DIMITRI VAN SCHAIK AN ANALYSIS OF MERGER WAVES AND HOSTILE TAKEOVERS The number of mergers between large companies has been low, and hostile takeover cases have been rare in post-war Japan. Since the 1990s, however, total M&A activity has been increasing in number of cases and in value. Coinciding with this trend, hostile tender M&A in Japan

offer attempts are also more frequent. Previous research argues that the low level of - M&A in Japan DIMITRI VAN SCHAIK merger and hostile takeover activity is caused by three institutional elements within the An Analysis of Merger Waves and Hostile Takeovers Japanese society: the main bank system, the horizontal keiretsu, and the specific Japanese culture. With reference to hostile takeovers we show that it is important to make a distinction between greenmail and hostile tender offers. We build an institutional model that emphasizes the necessity to consider institutional elements from a historical and context- specific perspective. Our model indicates that the vertical keiretsu and the trade association have an important impact on hostile tender offers in Japan. Regarding mergers, we examine whether the main bank system influences merger activity of companies. By using two event studies we show that involvement of a main

ERIM bank does not create shareholder wealth in mergers. The main bank appears to act in order to protect its own interests as creditor. Design & layout: B&T Ontwerp en advies (www.b-en-t.nl) Print: Haveka (www.haveka.nl) Design & layout: B&T Ontwerp en advies (www.b-en-t.nl)

ERIM The Erasmus Research Institute of Management (ERIM) is the Research School (Onder - zoek school)in the field of management of the Erasmus University Rotterdam. The founding participants of ERIM are Rotterdam School of Management, Erasmus University, and the Erasmus School of Econo mics. ERIM was founded in1999 and is officially accredited by the Royal Netherlands Academy of Arts and Sciences (KNAW). The research undertaken by ERIM is focussed on the management of the firm in its environment, its intra- and interfirm relations, and its busi ness processes in their interdependent connections. The objective of ERIM is to carry out first rate research in management, and to offer an advanced doctoral pro gramme inResearch in Management. Within ERIM, over three hundred senior researchers and PhD candidates are active in the different research programmes. From a variety of acade mic backgrounds and expertises, the ERIM commu nity is united in striving for excellence and working at the fore front of creating new business knowledge. Erasmus Research Institute of ManagementErasmus Research -

ERIM PhD Series Research in Management

Rotterdam School of Management Tel. +31 10 408 11 82 ERIM Erasmus University Fax +31 10 408 96 40 Erasmus School of Economics E-mail [email protected] P.O. Box 1738, 3000 DR Rotterdam Internet www.erim.eur.nl The Netherlands M&A in Japan An analysis of merger waves and hostile takeovers

1 2 M&A in Japan An analysis of merger waves and hostile takeovers

M&A in Japan Een analyse van fusiegolven en vijandige overnames

Thesis

to obtain the degree of Doctor from the Erasmus University Rotterdam by command of the rector magnificus Prof.dr. S.W.J. Lamberts and in accordance with the decision of the Doctorate Board.

The public defense shall be held on Thursday, 30 October 2008 at 13.30 hours

by Dimitri van Schaik born in Sliedrecht, the Netherlands.

3 Doctoral Committee

Promotores: Prof.dr. J. Spronk Prof.dr. J.P.M. Groenewegen

Other members: Prof.dr. G.A. van der Knaap Prof.dr. H.T.J. Smit Prof.dr. J.A. Stam

Copromotor: Dr. O.W. Steenbeek

Erasmus Research Institute of Management (ERIM) RSM Erasmus University / Erasmus School of Economics Erasmus University Rotterdam

Internet: http://www.erim.eur.nl

ERIM Electronic Series Portal: http://hdl.handle.net/1765/1

ERIM PhD Series in Research in Management, No. 141

ISBN 978-90-5892-169-7

All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means electronic or mechanical, including photocopying, recording, or by any information storage and retrieval system, without permission in writing from the author.

4 Contents

List of Figures iv List of Tables iv Preface vii

1 Introduction, Findings and Overview 1 1-1 Research question 3 1-2 Findings 5 1-3 Overview 7

2M&AactivityinJapan 11 2-1 Pre-war mergers 12 2-1-1 TSE listed company data 14 2-2 Post-war mergers 19 2-2-1 First merger wave 21 2-2-2 Latter half 1980s 25 2-2-3 Second merger wave 25 2-3 Hostile takeovers 32 2-4 Hostile tender offer attempts 35 2-5 Conclusion 40

3 Informal and formal institutions 43 3-1 Informal institutions 44 3-1-1 Religion and tradition 44 3-1-2 Constitution of Seventeen Articles 46 3-1-3 Ie system 48 3-2 Formal institutions 50 3-2-1 Pre-war political development and financial markets 50 3-2-2 Post-war political development and financial markets 53 3-3 Formal institutions related to M&A 56 3-3-1 Company Law 56 3-3-2 Antimonopoly Act 61 3-3-3 Securities Exchange Law 62 3-3-4 Other laws related to M&A 62 3-4 Summary 63

5 4 Ownership and corporate governance 65 4-1 Economic characteristics 66 4-1-1 Economic development 66 4-1-2 Administrative guidance (gyǀsei shidǀ)71 4-2 Pre-war financing and ownership 74 4-2-1 Pre-war zaibatsu 77 4-2-2 Pre-war corporate governance 78 4-3 Post-war financing and ownership 79 4-3-1 Post-war financing decision 82 4-3-2 Post-war keiretsu 83 4-3-3 Post-war corporate governance 84 4-3-4 Recent changes in corporate governance and ownership 86 4-4 Summary 89

5 Mitsubishi groups 91 5-1 Mitsubishi zaibatsu 92 5-2 Mitsubishi horizontal keiretsu 96 5-2-1 Data and Methodology 96 5-3 Mitsubishi vertical keiretsu 102 5-4 Conclusions 104

6 Institutional model for M&A 105 6-1 Framework of institutional model 106 6-2 Informal and formal institutions 109 6-3 Governance structure and M&A activity 111 6-3-1 Transaction cost economics 112 6-4 Actors 114 6-5 Institutional change 119 6-6 Concluding remarks 121

7 Institutional approach to hostile takeovers 123 7-1 Interpretations from previous research 124 7-1-1 Horizontal keiretsu 124 7-1-2 Culture 125 7-1-3 Transaction cost economics and vertical keiretsu 126 7-2 Corporate groups according to our model 127 7-2-1 Trust 128 7-2-2 Horizontal keiretsu 129 7-2-3 Vertical keiretsu 132 7-2-4 Trade associations 134 7-3 Hostile takeovers according to our model 136 7-3-1 Hostile tender offer attempts 138 7-3-2 Generated hypotheses 141 7-4 Concluding remarks 141

6 8 Introduction to M&A 143 8-1 M&A terminology 144 8-1-1 Types of M&A 144 8-1-2 Mergers 146 8-1-3 Tender offers 148 8-1-3-1 Defensive measures 148 8-2 Merger motives 149 8-3 Merger waves 152 8-3-1 Drivers for merger waves 152 8-3-2 Merger waves in the United States 153 8-4 Conclusion 156

9 Merger activity 159 9-1 M&A activity over time 160 9-2 Previous research 163 9-2-1 Main bank monitoring 163 9-2-2 Previous event studies 164 9-2-3 Profitability drivers 166 9-3 Shareholder wealth of bidder companies 169 9-3-1 Data-set 169 9-3-2 Results 174 9-3-3 Conclusions 186 9-4 Shareholder wealth of the bidder and target companies 187 9-4-1 Data 188 9-4-2 Results 194 9-4-3 Conclusions 205 9-5 Conclusion 205

10 Summary and conclusions 211 10-1 Summary 211 10-2 Conclusions 216

Dutch summary 221

References 225

About the author 237

ERIM PhD Series 239

iii

7 List of Figures

2-1 Mergers in the period 1906-1937 15 2-2 Mergers in the period 1950-1997 20 2-3 Mergers between large companies in the period 1957-1978 24 2-4 M&A cases in the period 1985-2006 30 4-1 GNP growth in the period 1886-1942 68 4-2 GDP growth and inflation rate in the period 1952-2004 70 4-3 Financing as percentage of GNP in the period 1903-1944 76 4-4a Nikkei 225 daily stock average price index 80 4-4b Share ownership in 1945, 1949-53 80 4-5a Stockownership 1950-75 (%) 81 4-5b Stockownership 1950-75 (shares) 81 4-6 Loan/assets-ratio of manufacturing companies in period 1960-2006 87 5-1 Loans from Mitsubishi Bank and Mitsubishi Trust Bank in 1975 98 5-2 Mitsubishi Keiretsu Diamond 100 6-1 Institutional model for M&A 107 6-2 Simple contracting schema 114 7-1 Ownership transformation of Mitsubishi groups 131 8-1 M&A types 145 8-2 Mergers in the United States in the period 1895-1930 153 8-3 M&A announcements in the United States in the period 1963-1998 154 9-1 CAR results of previous event studies on Japanese mergers 165 9-2 Inter-industry vs. intra-industry deals 174 9-3a CARs Japanese bidders 175 9-3b Excess trading volume Japanese bidders 175 9-4 Merger activity and Nikkei Index in the period 1981-2003 189

List of Tables

2-1 Sample companies by mergers and zaibatsu in the period 1906-1937 16 2-2 Mergers by size of bidder and target in the period 1906-1937 17 2-3 Mergers by industrial sector in the period 1916-1933 18 2-4 Mergers by size in sub-periods of 1953-1976 21 2-5 Mergers by industrial sector in sub-periods of 1953-1976 22 2-6 Merger activity by number of cases and by amount in the period 1984-2006 26 2-7 Comparison of merger activity by industrial sector between 1984-1990 and 1991-1997 27 2-8 Mergers by industrial sector in the period 1991-1997 28 2-9 Mergers by merger type in the period 1977-1983 29 2-10 M&A activity categorized by in-group and out-group cases in the period 1990-2006 31 2-11 Greenmail cases in the 1970s and 1980s 33

8 2-12 Hostile tender offers in Japan 35 3-1 Important laws and regulations related to M&A as of 1997 56 3-2 Ownership and rights in post-war Japan 58 4-1 Government policy plans for the period 1955-1971 72 4-2 Sources of funds for companies in the period 1931-1944 75 4-3 New share issues in the period 1956-1990 82 4-4 Net sources of funds for Japanese non-financial companies 82 4-5 Corporate loans by financial institutions in the period 1977-1991 85 4-6 Cross-shareholdings and long-term shareholdings in period 1987-2002 88 5-1 Ownership of Mitsubishi zaibatsu companies 95 5-2 Loan provision and shareholdings by Mitsubishi keiretsu financial institutions 97 5-3 Loan correlation Mitsubishi keiretsu financial institutions 98 5-4 Group ownership in Mitsubishi keiretsu 99 5-5 Ownership of/by diamond companies and financial institutions 101 5-6 Cross-shareholdings of diamond companies and financial institutions 102 5-7 Shareholdings in Mitsubishi vertical keiretsu in the year 1988 103 8-1 Tender offers in the United States in the period 1980-1993 156 9-1 Descriptive statistics 170 9-2 Summary Merger Specific and Company Specific Statistics 173 9-3 CARs and Excess Volume for Japanese bidders 176 9-4 CAR [-1, 0] for Japanese bidders categorized by characteristics of bidder and transaction 178 9-5 EV [0, +1] for Japanese bidders categorized by characteristics of bidder and transaction 180 9-6 Regression analysis CARs and Excess Volume 184 9-7 Characteristics by period 190 9-8 Descriptive statistics of sample (in million yen) 192 9-9 ROA data of bidders and targets 194 9-10 CARs for Japanese targets and bidders in the period 1982-2003 195 9-11 CAR [-1, +1] for Japanese bidders and targets categorized by characteristics of the merging companies 197 9-12 CAR [-1, +1] for Japanese bidders and targets categorized by main bank and financial distress 198 9-13 CAR [-1, +1] for Japanese bidders and targets by period and main bank 201 9-14 Classification of merging companies by financial health 203 9-15 Target in financial distress 204 9-16 CAR [-5, +90] for Japanese bidders categorized by characteristics of bidder and transaction 208

9 vi

10 Preface

I started working on this project in 2003, after receiving a Monbukagakusho scholarship that enabled me to go to Japan as a research student. It was a great privilege to be allowed to conduct my research at the Hitotsubashi University in Tokyo. The wealth of information available in the university’s library, combined with the input from the friendly and interesting people I met, has been of invaluable importance to the conceptualization of this book. I did most of my research in Japan and returned to the Netherlands in 2007 for the finalization of my thesis. In this preface I would like to express my gratitude to some of the many people without whom the finalization of my thesis would not have been accomplished.

First of all, I would like to thank the supervisors of my thesis for advising and inspiring me: First, Jaap Spronk, professor of Finance at the Erasmus University Rotterdam, for his devotion to my project. His directness and optimism have been of major importance to me, and our discussions have enabled me to structure and incorporate into this book the many thoughts and ideas I accumulated during my years in Japan. Second, John Groenewegen, professor of Institutional Economics at Delft University, who motivated me and guided me in exploring and benefiting from the doctrines of institutional economics. Not only has institutional economics become an important part of my thesis, I am convinced that the new ideas and insights I gained will be very useful for me in the future. I enjoyed the discussions on the institutional model for M&A as he always offered me new views. Third, Onno Steenbeek, doctor in Finance at the Erasmus University Rotterdam, for his assistance during the entire process of writing the thesis. Starting this project together in 2003, I am very grateful for his continued support and his willingness to act as my copromotor. I would like to express my thanks to the members of the inner committee, professors Van der Knaap, Smit, and Stam, as well as the other committee members, for reading and reviewing my thesis.

While in Japan I have met many people who also have been very important for the completion of this research, and, in particular, I would like to thank the following persons: First, professor Kitamura of Hitotsubashi University, for acting as my promoter for the Monbukagakusho scholarship and for providing me help related to my research during my years in Japan. Second, professor Vikas Mehrotra of the University of Alberta, for our lunches in Kunitachi and his willingness to listen to my new and different thoughts on (accepted) theories about Japan, providing me with new issues to consider. I also enjoyed our joint research.

vii

11 Third, I have had the privilege of meeting many renowned professors who always provided me with new ideas. In particular, I would like to thank professor Fukao (Hitotsubashi University), professor Teranishi (Nihon University), professor Okazaki (Tokyo University), and professor Miyajima (Waseda University) for taking the time to meet me and discuss the findings of my thesis when I visited Japan in February of this year.

Last but not least, I would like to express my thanks to some persons that provided me invaluable support on a personal level: First, my parents, for supporting me throughout the entire process of writing this thesis. Without their love and support I would not have been able to finish the book. Second, my very dear friend Takako, for providing me with a lot of encouragement and support while I was working on the research in Japan. Third, Erik, one of my best friends, for his continued moral support in the ups and downs of this project, and his opinion on subjects ranging from the thoughts of Aristotle to the design of the cover of this book.

Utrecht, August 2008

viii

12 1

Introduction, Findings and Overview

The number of mergers between large companies has been low, and hostile takeover cases have been rare in post-war Japan. Since the 1990s, however, total merger and acquisition (“M&A”) activity is increasing in both number of cases and in value. Coinciding with the trend in total M&A activity, hostile tender offer attempts are also more frequent with thirteen cases in the period from 1999 to July 2007. The collapse of the bubble economy and the subsequent recession in the 1990s forced Japanese companies to question the efficiency of the traditional Japanese system of corporate governance, which was characterized by an important role for main banks. Although this corporate governance may have been beneficial for the production needs and technology of Japanese companies up to the 1990s, “those needs have changed and the main bank has largely ceased to function” (Milhaupt and West 2001, p.12). Deregulation of financial markets has reduced the influence of the main bank, and the banking crisis and new accounting rules in the latter half of the 1990s have resulted in unwinding of stable and cross-shareholdings. Specific Japanese cultural characteristics that are claimed to have limited M&A have changed as well, and major legislative amendments regarding M&A were announced and implemented from the late 1990s onward. Lin (2005) notes that the lifetime employment system has become increasingly difficult to sustain for Japanese companies, and “the keiretsu relationships in Japan were drifting away from the keiretsu governance modes towards the American style of arms-length contracting and top-down administration” (p.98). These factors have made it easier for companies to engage in M&A activity and created various incentives to restructure operations. This thesis examines how and why M&A activity in Japan has changed over time. Previous research argues that the low level of merger and hostile takeover activity is caused by three institutional elements within the Japanese society: the main bank system, the horizontal keiretsu, and the specific Japanese culture. 1 Two other important

1 These two post-war characteristics will be discussed in detail in Chapter 5 and 6.

13 Chapter 1 Introduction, Findings and Overview

institutional elements, the vertical keiretsu and trade association, have not been included in the M&A discussion. This thesis shows that these two latter institutional elements need to be considered, as specifically these factors may have played an important role in hostile takeover activity in Japan. The main bank system defines the special relationship that exists between a company and its most important (main) bank. The involvement of the main bank in the operations of the company is through bank loans, cross-shareholdings, payment settlement accounts, information services, supply of management resources, and bond-issue related services (Aoki et al. 1994). Through the special relationship with companies, the main bank has a lot of information about its customers and also functions as monitor and advisor of the company. Aoki et al. (1994) argue that corporate monitoring by the main bank substitutes the market for corporate mergers and takeovers in post-war Japan. The horizontal keiretsu (kigyǀ shǌdan) is a horizontally diversified industrial group in which companies have ties with banks, trading companies, customers, and other affiliated companies. The ties in the industrial group are characterized by the stable and cross- shareholdings and implicit contracts (Sheard 1989; Aoki and Patrick 1994; Nakatani 1984).2 Hostile takeovers are argued to be prevented by stable and cross-shareholdings within the horizontal keiretsu (Morck and Nakamura 2003; Sheard 1989; Aoki and Patrick 1994; Nakatani 1984). According to the Japanese cultural characteristics explanation, the Japanese have a unique capacity to cooperate with each other in harmony and they put high emphasis on trust (Johnson 1982). Merger and hostile takeover activity is argued to be low because Japanese companies are unwilling to do so. This unwillingness derives from various characteristics within the Japanese (corporate) culture. Most important is that a Japanese company is regarded as a community that workers identify their interests with, as a place to both work and live. This resulted in cultural characteristics such as lifetime employment, internal promotion, restricted labour mobility, seniority-based wages, and company unions that are organized within and oriented to the company (Pettway and Yamada 1986; Odagiri and Hase 1989; Abegglen 1983). Because the employees are most important in Japanese companies, a company’s management is unwilling to accept a merger, and hostile takeovers will be strongly resisted. The importance of the employees is further reflected in the preference for growth through internal corporate growth over M&A as this can contribute to the use and enrichment of existing human resources and create promotion opportunities for employees. Odagiri and Hase (1989) point out that Japanese companies prefer capital participation and acquisitions to mergers. They argue that two different companies have to be unified in a merger, and this is rather avoided because it could cause various labour-related problems. Based on the above-mentioned arguments, we conclude that in order to understand merger and hostile takeover activity in Japan, we need to (i) be aware of the ownership structure and group structure of Japanese companies, and (ii) comprehend how Japan’s informal and formal institutions influence the ownership structure, group structure, and M&A activity. Regarding hostile takeovers we argue that it is necessary to make a distinction between hostile tender offers and greenmail. The difference between these two activities is the motivation of the bidder (raider). In a hostile tender offer the bidder wants

2 Aoki et al. (1994) indicates that although not all companies are part of a kigyǀ shǌdan, almost every Japanese company has a main bank.

14 Section 1-1 Research question

to attain operational control over the target company. He tries to obtain an ownership percentage that is sufficient to exercise influence over managerial and financial decisions. The strength of the bidder’s influence depends on the actual ownership percentage he obtains. In greenmail, on the other hand, the bidder does not have the intention to gain control. Greenmail is the practice of a target company buying back its own shares from a hostile bidder at a substantial premium, herewith getting rid of the hostile threat.

1-1 Research question In the M&A process the most important directly involved actors are the management, shareholders, board of directors, and creditors. Their relative influence primarily depends on the company’s financing decision: the management’s choice between internal, debt, and equity financing. The financing decision results in a company’s capital structure and has a determining effect on the ownership structure3 and corporate governance structure. Other actors, with an indirect effect on M&A, are the government, employees, and other stakeholders. There are various factors that influence M&A activity, such as problems and/or opportunities caused by alterations in economic conditions, efficiency conditions, and/or technology. It can also be related to industrial development, government policy and regulations, or (international) competition. The financing decision of companies depends on informal and formal institutions, and is influenced by the historical and context-specific background of managers and companies. In this thesis we therefore look into the economic development as of 1868, when the Meiji government was established. 4 We examine how changes in laws and regulations, in relation to both the economic development and informal institutions, influence the financing decision, ownership structure, corporate governance structure, and M&A activity. Including the pre-war period and post-war period in the analysis is interesting because Japan faced new formal institutions in both periods. In the pre-war period these were implemented after she opened her ports and had to implement her first constitution and various laws to modernize the country. The second exogenous change in formal institutions was when U.S. occupation forces designed and implemented a new constitution and various new laws and regulations after the Second World War (“WWII”). Especially the new laws after WWII had a large effect on the financial markets, formation of corporate groups, and companies’ financing decision, ownership structure, and corporate governance structure. This thesis investigates these effects and subsequently looks into their influence on M&A activity. We pay special attention to the changes in ownership and group structures between the pre-war period and post-war period. With reference to the post-war period we examine the role of the main bank in detail.

3 The influence of the financing decision on the ownership structure is through private placement of shares. 4 Meiji period starts in 1868 en ends in 1912.

15 Chapter 1 Introduction, Findings and Overview

Central question What are the determinants of M&A activity and merger waves in Japan over time?

Sub-question I Is hostility in M&A activity increasing, and which actors initiate hostile takeover activity?

Sub-question II What motivates Japanese companies to engage in mergers, which actors play a role, what are the profitability drivers, and is shareholder wealth created?

The thesis looks into M&A activity in Japan over time, examining the pre-war period and post-war period. M&A generally refers to a broad range of M&A structures including capital participations, business tie-ups, divestures, and other forms of restructuring. We primarily focus on mergers and hostile takeovers, the latter defined as greenmail and hostile tender offers. We attempt to find the motives for companies to engage in mergers, and the drivers for Japanese merger waves during the period under study. After a general overview of M&A activity, we describe aspects in the Japanese economy that (can) play an important role in M&A. Subsequently, we build an institutional model that integrates these aspects and indicates their interrelations and influence on M&A activity. Our approach to sub-question I is to analyze hostility in M&A activity with a qualitative study. Our decision for a qualitative study is based on the number of hostile takeover cases, the state of current knowledge in the field of hostile takeovers in Japan, and the objective of our thesis. We use the theory of social ontology that “indicates the lines of inquiry required to produce a complete description. If we start research by describing the nature of social phenomena as they are experienced, it will make a difference in structuring data gathering; in developing a research craft capable of seeing practice, interaction manouvres, and tacit embodiment; in shaping a research agenda; and, ultimately, in where we end substantively” (Katz 2002, p.255). Previous research on hostile takeovers in Japan is limited, and the existing literature provides only two explanations for the low number of cases: stable and cross-shareholdings and the specific Japanese culture. At this stage the issues that influence hostile tender offers and greenmail in Japan have not been clearly identified. By using an inductive and hypothesis-generating methodology, we aim at gaining a clearer picture and understanding. We adopt an interpretive research paradigm and, following an inductive research path, we build a model inspired on insights from institutional economics. The interpretive methodology can be beneficial in identifying unknown variables, themes and processes that may be valuable in developing an explanatory theory in the area under study. Institutional economics is particularly applicable as it provides a holistic research approach, and allows an inductive generation of a theory from the available empirical data. The model puts the emphasis on the historical and context-specific background, which allows us a dynamic investigation of institutional elements. According to institutional economics, the historical development of institutions and the context-specific background have an important effect on new institutions. Our analysis therefore includes the cultural and social world that members of a society share and internalize. The model combines the various elements that influence M&A activity, by integrating the informal and formal institutions, ownership structure, and group structure of companies. We not only look into the main bank system, horizontal keiretsu and Japanese culture, but

16 Section 1-2 Findings

also the vertical keiretsu and the trade association. The vertical keiretsu is a group of companies comprising of a parent company and subsidiaries; the subsidiaries take part in the production process of the parent company. A trade association is a group formed by companies with a common business interest. In our institutional analysis we examine the historical development of the corporate groups and analyze their influence on hostile takeover activity. Regarding sub-question II, we follow a deductive research path and empirically test whether domestic mergers in Japan create shareholder wealth. We conduct two event studies to examine the explanatory power of the previously mentioned institutional elements within the Japanese economy, such as the main bank, horizontal keiretsu, and vertical keiretsu.

1-2 Findings We assemble a unique data-set on M&A activity in pre-war and post-war Japan. For the pre-war period we examine 537 companies listed on the Tokyo Stock Exchange and find that M&A activity is very low in the period 1906-37. We do not find evidence that a lot of hostile takeovers occurred, and companies related to a zaibatsu group were not more active in M&A than companies without relations with a zaibatsu group. In the post-war period we find two merger waves; the first wave occurs in the period 1963-72, and the second wave starts in 1991 and continues up to present. The first merger wave coincides with trade liberalization and removal of controls on capital transactions in 1964. These mergers were aimed at preventing excessive competition between domestic companies and increasing their international competitiveness. The second wave starts after the collapse of the bubble economy and is related to corporate restructurings. After the regulatory changes in the latter half of the 1990s, the number and size of total M&A activity increases considerably, but the number of mergers remains relatively low and stable in the period 1999-2006. In contrast to the pro-cyclical merger waves in the United States, we find that mergers in Japan tend to be counter-cyclical, both with respect to the general economy as well as with respect to stock market valuations. With reference to hostile takeovers in the post-war period, we indicate that it is important to make a distinction between greenmail and hostile tender offers. For both types of hostile takeovers we discuss some cases and we show that the first hostile tender offer was initiated in 1999. With our institutional model we look into detail in hostile takeovers in post-war Japan. Previous research argues that the low level of hostile takeover activity is primarily caused by two institutional elements in the Japanese society. The first is the horizontal keiretsu, which is argued to have prevented hostile takeovers with stable and cross-shareholdings between companies. The second is the specific Japanese culture that is said to be characterized by the pursuit of harmony. Above we indicated that cross-shareholdings are unwinding and Japanese cultural characteristics changing. Although we see a higher level of hostile tender offers, there was only one initiated by a Japanese blue-chip company. This raises the question whether the said institutional elements are indeed changing, or whether they do not have the influence as claimed in previous research. This thesis provides evidence that the influence of the vertical keiretsu and trade associations can also be important. These two institutional elements have been neglected in the hostile takeover discussion, but could be indispensable to understanding it and making predictions on its development in the future.

17 Chapter 1 Introduction, Findings and Overview

We show that the claim that horizontal keiretsu is the most important prevention measure against hostile takeovers is doubtful. New laws implemented after WWII created the possibility to engage in greenmail and, as companies could not repurchase their own shares prior to 2001, companies that were part of a horizontal keiretsu were targeted. Rather than preventing hostile takeovers, the horizontal keiretsu appear to have caused greenmail in the post-war period. The argument that the Japanese, as a people, put high emphasis on harmony and trust implies that Japanese would not undertake hostile actions. Informal institutions change slowly and we do not find evidence that these have changed to a high degree in recent years. We show that trust needs to be divided into general trust and particularistic trust. General trust is trust between individuals that do not have a particular relationship, and particularistic trust involves trust in groups of mutually dependent individuals. The Japanese society is characterized by low general trust and high particularistic trust. The low general trust can be seen in the greenmail practices towards companies, and the high particularistic trust in, for example, trade associations. According to our institutional model we conclude that the nature of hostile takeovers in Japan is not changing, as the recent hostile tender offer attempts can (also) be interpreted as greenmail. This, combined with the fact that attempts remain at a very low level, leads us to question the generally accepted above-described explanations and the influence of the unwinding of cross-shareholdings and changes in Japanese cultural characteristics. With our model we therefore propose the trade association and vertical keiretsu as explanations why Japanese blue-chip companies do not initiate hostile tender offers. Trade associations are formed by companies in the same industrial sector, characterized by information exchange and strong relationships. We show that because of the relationships in these groups, companies built a high level of particularistic trust, preventing these companies to engage in hostile takeovers against each other. The vertical keiretsu is another important impediment to hostile takeovers, functioning in three ways. The first characteristic is that a large Japanese company is not the same as a large U.S. company. Whereas U.S. companies tend to produce most products in-house, the Japanese company uses a wide subcontracting system that makes it a difficult target for a hostile takeover. The second characteristic of the vertical keiretsu is that supply from the subsidiaries is indispensable in the production process of the parent company. The majority of its important subsidiaries is already fully owned, but if an important partially owned subsidiary is attacked with a hostile takeover, the parent company will intervene to secure its own production process. The third aspect is the question whether it would be beneficial for a hostile bidder to attempt a hostile tender offer of an unwilling subsidiary company. Even if the hostile bidder would succeed, it is very likely it will lose the subsidiary’s suppliers and customers in the process. All these factors have limited hostile takeover attempts to a high degree. Regarding the mergers in the second wave we investigate whether shareholder wealth is created. We conduct two event studies and examine the influence of some institutional elements, specifically the main bank, horizontal keiretsu and presence of a parent company, and also include company-related data such as profitability and financial distress. In our first event study we examine the announcement returns of listed bidder companies, but we do not find a positive association with these returns and the bidder’s affiliation with a main bank or horizontal keiretsu. When the bidder and target company have the same main bank, its involvement does not create shareholder wealth, but is motivated to protect its own

18 Section 1-3 Overview

interests as creditor; same main bank mergers are not between two weak companies, but at least one of the merging companies is financially strong. In the early 1990s most mergers occur between a weak target company and strong bidder company.

1-3 Overview This thesis consists of 10 chapters, divided into 4 parts. Chapter 1, as an introductory chapter, explains the research question and provides the findings and structure of the thesis. Part I consists of chapter 2 and presents an overview of M&A activity in Japan over time. Chapters 3 through 5 form Part II and provide background information for interpretation of M&A activity in the pre-war period and post-war period. Part III, chapters 6 and 7, introduces our institutional model and examines hostile takeover activity, and Part IV, consisting of chapters 8 and 9, focuses on mergers. Chapter 10 summarizes and concludes the thesis.

Part I In Chapter 2 we start with a description of M&A activity in Japan during the pre-war and post-war period. In particular, we examine whether merger waves occurred and start the analysis of both periods at a macro-level. If we find a period with increased merger activity, we turn the analysis to the meso-level. In this chapter we also discuss hostile takeovers in the post-war period and describe the most important greenmail cases and hostile tender offer attempts.

Part II Chapter 3 discusses informal institutions, political development, and formal institutions. Informal institutions are traced back to the origins of Shinto, Buddhism, and Confucianism. Their influence on the (moral) Constitution of Seventeen Articles and the ie system is also investigated. We look into political development as of 1639, when Japan implemented a political system that banned almost all foreign contact, and we describe laws related to financial markets and M&A activity. In Chapter 4 we examine the influence of the informal and formal institutions on the financing decision, ownership structure, and corporate governance structure of Japanese companies. After a brief economic overview of the pre-war and post-war periods, in which special attention is given to the role of the government and trade associations, we look into the financing decision in these periods with aggregate data. Next, we look into the ownership structure and corporate governance structure. We also examine corporate groups: (i) zaibatsu for the pre-war period, and (ii) the horizontal keiretsu, main bank system, and vertical keiretsu for the post-war period. Chapter 5 presents a case study of the Mitsubishi groups. Mitsubishi is taken as an example because it originated at the start of the Meiji period (1868-1912) and became the second largest zaibatsu. In the pre-war period we look into the Mitsubishi zaibatsu group, and in the post-war period the Mitsubishi horizontal keiretsu, the Mitsubishi vertical keiretsu, and the main bank system with the Mitsubishi Bank at the apex. We pay special attention to the financing decision and resulting ownership structure and corporate governance structure in the two periods. In chapter 7 we revisit these findings and examine the transformation from Mitsubishi zaibatsu to Mitsubishi horizontal keiretsu with our institutional model in order to interpret their influence on hostile takeovers.

19 Chapter 1 Introduction, Findings and Overview

Part III Chapter 6 introduces the institutional model that we build to analyze M&A activity in Japan. Our model is based on the Williamson models and uses insights from North and Greif. The model emphasizes the importance of motivation of actors and the historical and context-specific background of institutional elements. Regarding the governance structure of companies, we pay special attention to transaction cost economics and institutional change. M&A activity is a strategic decision by individual actors and influenced by all levels of the model and the governance structure. At the same time, M&A activity has an important impact on individual actors and governance structure as well. Chapter 7 analyzes hostile takeovers in Japan according to our institutional model. We examine the influence of changes in the legal system and specific cultural characteristics, such as trust. As corporate groups are important in the Japanese economy, we first use the model to analyze the following corporate groups: horizontal keiretsu, vertical keiretsu, and trade association. Next, we use our findings on the corporate groups and trust to explain why the number of hostile takeovers has been low in post-war Japan.

Part IV Chapter 8 provides an overview of the various types of M&A and explains merger motives and drivers for merger waves. We discuss mergers and other types of M&A activity. Regarding hostile tender offer bids, we indicate some available defensive measures. The chapter concludes with an overview of the five merger waves that occurred in the United States. In Chapter 9 we interpret the merger activity and merger waves as discussed in chapter 2. We look at the merger motives and drivers for merger waves, and integrate political and economic development. Results from previous studies on mergers and the monitoring role of the main bank are discussed as well. We execute two event studies and examine whether shareholder wealth is created around the announcement date of a merger. In this analysis we look into the influence of various institutional elements in Japan, such as the main bank, the horizontal keiretsu, (changes in) laws, and economic development. In the first research we look into shareholder wealth of bidder companies between 1993 and 2003 in 136 domestic mergers between non-financial companies. The sample includes listed target companies as well as unlisted target companies. In the second research we examine the abnormal return of listed bidder companies and listed target companies in 91 mergers between non-financial companies in the period 1981 to 2003.

20 Section 1-3 Overview

1 Introduction, Findings & Overview Research question Findings Overview

2 M&A activity in Japan Pre-war Mergers Post-war Mergers I Greenmail Hostile tender offer attempts

3 Informal and formal institutions Informal institutions Political development Formal institutions

4 Ownership and corporate governance Economic development Pre-war financing, ownership structure, and corporate groupings II Post-war financing, ownership structure, and corporate groupings

5 Mitsubishi groups Mitsubishi zaibatsu Mitsubishi horizontal keiretsu Mitsubishi vertical keiretsu

6 Institutional model for M&A Framework of institutional model Transaction cost economics Institutional change IIIIV 7 Institutional approach Interpretations from previous research Corporate groups according to our model Hostile takeovers according to our model

8 Introduction to M&A Types of M&A Mergers motives Drivers for merger waves U.S. merger waves IV 9 Merger activity Interpretation merger waves over time Main bank monitoring Shareholder wealth of listed bidder companies Shareholder wealth of listed bidder and listed target companies

10 Conclusion

21 10

22 2

M&A activity in Japan

This chapter describes M&A activity in Japan over time, covering the pre-war period and post-war period. We compile a database that, to our knowledge, contains the most extensive information on Japanese M&A activity in these periods compared to previous research. Section 2-1 presents M&A activity of 537 companies listed on the Tokyo Stock Exchange in 1937. In section 2-2 we present domestic merger data from the annual reports of the Japanese Fair Trade Commission for the period 1949-97, and we use information from Recof5, a company that collects data on M&A transactions since 1985, for the period 1998-2006. Hostile takeover activity in the post-war period, defined as greenmail and hostile tender offers, is discussed in sections 2-3 and 2-4. The final section summarizes and concludes the chapter.

5 http://www.recof.co.jp/english/

23 Chapter 2 M&A activity in Japan

2-1 Pre-war mergers R M&A activity is low in the pre-war period. R Most M&A transactions occur in the 1920s, and they are concentrated in the electricity sector. R Companies affiliated with a zaibatsu group are not more active in M&A activity than companies that are not affiliated with a zaibatsu group.

In this section and section 2-2 we begin with a discussion of merger activity at a macro- level and investigate whether there are periods with increased merger activity. If we find a higher level of merger activity in a particular period, we look into the meso-level and investigate if the period can be termed as a merger wave. We use the term ‘merger wave’ according to Nelson’s definition of ‘merger movement’. Nelson (1959) argues that “[I]f between two successive periods in time, merger activity in a number of different industries changes in the same direction, we have a sound basis for characterizing the more active period as what is usually called a merger movement” (p.38). The period of increased merger activity is particularly interesting to examine in comparison with the preceding period of lower merger activity. This provides an ideal situation to investigate the merger wave drivers. We subsequently look into the influence of various factors that can influence mergers such as problems and/or opportunities caused by alterations in economic conditions, efficiency conditions, and/or technology. Merger activity can also be related to industrial development, state policy, government regulations, or (international) competition. In our examination of merger activity during the pre-war period from 1868 to 1937, we first discuss findings from previous research and then turn to our own comprehensive analysis. The research on M&A in the pre-war period is very limited, but two periods of increased M&A activity can be distinguished from previous research: the first period is from 1882 to 1885, and the second period is the 1920s. As a result of the Matsukata reforms in the 1880s, a lot of companies went bankrupt and the number of companies fell from 3,336 in 1882 to 1,279 in 1885. As the size of the companies in 1885 had increased considerably, Röhl (2005, p.346) assumes that a lot of the failing companies were absorbed by healthier companies. The 1920s, characterized by the many financial crises and the Kantǀ earthquake, caused a lot of companies to fall into bankruptcy and a similar trend as in the 1880s occurred. Previous research provides anecdotal evidence of companies that failed in the 1920s, including zaibatsu companies such as Kuhara and Fuijita. Flath (2000) mentions that the “larger zaibatsu6 emerged from the war boom stronger than ever, and proceeded to absorb many of the failing businesses” (p.52). Nakamura (1994) argues about the bankruptcies that “[T]he four big zaibatsu7 and the major spinning companies came through largely unscathed owing to their conservative management policies and so stayed ahead where other companies fell behind” (p.11). These developments had an important effect on industries’ concentration ratio and the size of surviving companies. The stronger companies were primarily the zaibatsu companies that increased their influence on the

6 “Larger zaibatsu” are Old zaibatsu with origins prior to or beginning in the Meiji period, such as Mitsui, Mitsubishi, Sumitomo and Yasuda. Zaibatsu groups such as Kuhara and Fujita are know as Taishǀ zaibatsu and emerged resulting from the economic boom of WWI. 7 Mitsui, Mitsubishi, Sumitomo and Yasuda

24 Section 2-1 Pre-war mergers

overall economy. Another characteristic of the 1920s is that a lot of banks failed and the total number fell from 2,036 banks in 1920 to 895 banks in 1930. Similar to the above- described process with regard to companies, stronger banks absorbed failing banks and herewith strengthened their own position. Based on these two periods of increased M&A activity in pre-war Japan, we see that these were primarily aimed at absorbing companies in financial distress (Moulton 1931; Aoyama and Nishikawa 1958). Okazaki (1999, 2001) looks into M&A activity of zaibatsu companies and examines the six largest zaibatsu groups. We discuss his findings below but first look into his detailed explanation of two M&A deals in pre-war Japan (his research covers four deals in total). These case studies provide insight into the major actors and motives of the mergers and we give short summaries of two deals:8 Tokyo Steel - Tokyo Steel was established in 1917 and started a relationship with the Mitsubishi zaibatsu through a 500 thousand yen loan with the Banking Department of Mitsubishi Goshi in that year. Due to declining prices and excess inventory of products and raw materials, Tokyo Steel went into deficit two years later. The company’s financial situation deteriorated and as of 1925 the company “whose paid-in capital was 2,000,000 yen, owed more than 2,000,000 yen to Mitsubishi Bank, besides 1,400,000 yen to other creditors” (Okazaki 2001, p.263). Mitsubishi Bank investigated the financial needs and future prospects of Tokyo Steel, and as the company was considered important to the industries of Japan, proposed a rescue loan. Although Mitsubishi Bank only had the intention to provide the rescue loan, the presence of a hostile takeover led to acquisition of shares. After the acquisition, “84.3% of the shares were held by Isaji Endo, who was the chief of the Section of Loan, Mitsubishi Bank, and had been dispatched as an auditor in 1928” (Okazaki 2000, p.15). Management of the company was done by professionals from Mitsubishi Iron Works. Nihon Flour Mill - Nihon Flour Mill was established in 1896 and came to have relations with Suzuki Shoten9 in the early 1920s. “[I]t became increasingly dependent on Suzuki as its financial condition deteriorated. In 1922, Suzuki became top shareholder with a 13.1% holding” (Okazaki 2001, p.264). Nihon Flour Mill faced increasing financial difficulties in the mid-1920s and tried to negotiate a merger with Nisshin Flour Mill in 1926. Nisshin Flour Mill abandoned the merger because Nihon Flour Mill’s debt level was higher than expected. Hereafter Suzuki arranged, through its intimate business relations with the Taiwan Bank, a rescue loan for Nihon Flour Mill. Subsequently, Nihon Flour Mill reduced its capital by 75% and the shareholding of Suzuki in Nihon Flour Mill increased to 72.8%. Two executives from Suzuki and one from Taiwan bank were installed and restructuring was started under their management. In the financial crisis of 1927, however, Suzuki and Taiwan Bank bankrupted and Nihon Flour Mill “turned to Mitsui & Co. (Mitsui Bussan), with which it had trading connections” (p.264). A new board of directors for Nihon Flour Mill, including chairman and executive director, was formed by Mitsui & Co. that became a dominant shareholder with 60% after buying the shares held by Taiwan Bank.

The case studies by Okazaki (2001) present various important points on M&A activity in the pre-war period. They indicate that the holding company was most important in making decisions related to M&A, and show that banks had an important advising role to the

8 Please refer to Okazaki (2001) for more detailed information. 9 A major trading company which rapidly grew during the World War I.

25 Chapter 2 M&A activity in Japan

holding companies, besides providing rescue loans. In his investigation of the role of zaibatsu companies in pre-war M&A activity, Okazaki (1999, 2001) uses a data-set consisting of 97 zaibatsu companies. In order to establish whether a company is taken over by a zaibatsu company, he compares the year the company was established with the year the company came to belong to the respective zaibatsu. He finds that by this measurement 19 companies were taken over by zaibatsu companies (20%). He argues that the M&A market was active and that the zaibatsu companies played an important role. He mentions that “zaibatsu companies also disciplined the non-affiliated companies in the capital market. The holding companies and the core affiliated companies frequently executed takeover, which contributed to restructure the targeted companies, through replacement of the directors, assistance of management” (Okazaki 2000, p.1). Okazaki (1999, 2001) focuses on M&A activity of zaibatsu companies, and below we review his findings by investigating M&A activity of both zaibatsu companies and non- zaibatsu companies. Although Okazaki (1999, 2001) uses the term “discipline” with reference to takeovers in the pre-war period, this should be interpreted as changing the management of weak companies by strong companies, not as hostile takeovers. 10 The increased merger activity was primarily caused by failures as a result of financial difficulties due to the economic recession and financial crises in the 1920s. As Okazaki (2001) indicates with his quotation of Mitsubishi Steel, failing companies were absorbed with a rescue motive: “Tokyo Steel will not recover unless substantial new funding is invested. However, because its business is indispensable to the industries of Japan, it would be a great loss for our nation to let it go bankrupt” (Mitsubishi Steel 1985 p.111, quote from Okazaki 2001, p.263). In the following sub-section we look into merger and takeover activity of TSE listed companies in 1937, and investigate whether the zaibatsu companies were more involved in these activities than non-zaibatsu companies.

2-1-1 TSE listed company data We look into pre-war merger activity of 537 industrial companies listed on the Tokyo Stock Exchange in 1937. We analyze the corporate history of these companies for merger and takeover activity and herewith provide to our knowledge the most extensive data-set on M&A activity of industrial companies in pre-war Japan. Prior research focused on aggregate data of companies (Röhl 2005), or a limited number of case studies (Okazaki 1999, 2001). Miyajima, Imajoh and Kawamoto (2007) examine M&A activity of 200 top non-financial companies, measured as total assets in fiscal year 1937. Our data-set covers more companies as we collect information on M&A activity of all listed companies, as described in the Kabushiki Gaisha Nenkan of 1937 (Joint-stock company Yearbook). For these companies we collect all available data on merger activity in the period 1906-1937: the information in the corporate history provides us with (i) the number of merger cases for each company and (ii) the size of the merger transactions by equity value.11 Analysis of the number of merger cases gives equal weight to all transactions and indicates the breadth of merger activity. The size of the merger transaction indicates which periods are dominated

10 Based on discussion with Professor Okazaki in February 2008. 11 A merger case is a merger transaction and does not clarify the number of target companies involved. With reference to the size of the merger transaction, this includes all target companies in a multi-company merger.

26 Section 2-1 Pre-war mergers

by large transactions and measures the depth (materiality) of sizable merger transactions (Bruner 2004). The size of the merger transaction is measured as the written-up capitalization of the bidder company’s equity capital at the time of the merger. This amount is a lower limit of the actual transaction amount of merger activity, because we only look at the TSE listed companies and we could not deduct the merger value for 25 acquisitions that were paid in cash. For these 25 acquisitions there was no change in the bidder company’s equity capital and we assigned a value of zero. Figure 2-1 shows merger activity in the period 1906-37 as the number of merger cases and the total transaction value. Although overall merger activity was low in the pre-war period, not exceeding 41 merger cases in any year, there are clearly periods in which we can see concentrated merger activity. In the periods 1920- 23 and 1927-29 the annual average of merger cases was 35 at a value of 150 million yen. In comparison, the number of merger cases was 8 at a total transaction value of 10 million yen in the period 1906-19.

45 300

40 250 35 rnato iei l ye mln in size Transaction 30 200

25 150 20 Number of cases 15 100 n

10 50 5

0 0

9 3 5 9 5 9 1 5 7 90 91 91 91 921 92 92 93 93 93 1907 1 1911 1 1 1917 1 1 1923 1 1927 1 1 1933 1 1 Total cases Amount mln

Figure 2-1 Mergers in the period 1906-1937 Notes The sample consists of 537 non-financial listed bidder companies that engaged in absorption mergers, consolidations and tender offers in the period 1906-37. The transaction size of the merger is measured as the written-up capitalization of the bidder company’s equity capital at the time of the merger. The number of merger cases is depicted on the left vertical axis and the transaction size on the right vertical axis. Source Kabushiki Gaisha Nenkan (Joint-stock company Yearbook) of 1937. The corporate history of 537 non-financial companies listed on the TSE in the year 1937 was analyzed on M&A activities.

We first look into the size of bidder companies and the role of zaibatsu groups in merger activity. The size of bidder companies is the equity capitalization in the year 1937. Table 2-1 shows the number of companies in the sample by size and whether they belong to a zaibatsu group. Companies with equity capital lower than 5 million yen and in the group 5- 10 million yen are 29% and 19% of the total sample, respectively. Most companies belong to the group with equity capital of 10-50 million yen, at 41% of the total sample. There are

27 Chapter 2 M&A activity in Japan

61 companies with equity capital of more than 50 million yen. Column 2 shows that 45% of all sample companies were involved in one or more mergers. Bidder companies engaged in mergers are concentrated in the groups with higher equity capital; 109 bidder companies in the 10-50 million yen group (50%) and 49 companies in the group of 50 million or more (80%). In the group of the smallest companies only 28% of the 155 companies are involved in mergers. Next, we look into the importance of zaibatsu companies in merger activity and define zaibatsu companies according to the lists in Okazaki (1999). The lists provide the more important companies that belong to zaibatsu groups as of 1937. As he also looks at other information sources12 and includes financial companies, his total of zaibatsu companies is higher than in our data-set. We follow Okazaki in the list regarding the zaibatsu groups Mitsui, Sumitomo, Yasuda, Asano and Nissan; of the 76 companies in these 5 zaibatsu groups, we have 39 companies in our own data-set. The companies we do not have in our data are 15 financial companies and 22 companies that do not have widely held shares. We use data of Shibagaki (1968) to examine the companies that are part of the Mitsubishi group and our data-set includes 26 companies. In our sample 63 companies (12%) are related to a zaibatsu, and these companies primarily belong to the groups of larger companies with 56 of the 63 companies having an equity capital of more than 10 million yen.

Table 2-1 Sample companies by mergers and zaibatsu in the period 1906-1937 Total Sample Zaibatsu companies All Merger Company size % % of (a) All % of (a) Merger % of (b) (a) (b) <5mlnyen1552944282112 5-10 mln yen 102 19 40 39 5 5 2 5 10-50 mln yen 219 41 109 50 41 19 17 16 >50mlnyen6111498015251122 All 537 100% 242 45% 63 12% 31 13% Notes The sample consists of 537 non-financial companies listed on the TSE in 1937, of which 242 companies engaged in absorption mergers, consolidations and tender offers in the period 1906-37. The size of the bidder companies is the equity capitalization as of 1937. Zaibatsu companies belong to one of the following groups: Mitsubishi, Mitsui, Sumitomo, Yasuda, Asano and Nissan Sources Kabushiki Gaisha Nenkan (Joint-stock company Yearbook) of 1937. The corporate history of 537 non-financial companies listed on the TSE in the year 1937 was analyzed on M&A activities. To categorize whether a company belonged to the zaibatsu groups Okazaki (1999) was used for Mitsui, Sumitomo, Yasuda, Asano and Nissan, and Shibagaki (1968) with reference to Mitsubishi zaibatsu.

The number of zaibatsu companies involved in a merger is 31, which is 13% of all companies engaged in mergers. As this percentage is in line with the 12% of zaibatsu companies in the total sample, we can conclude that, as far as the number of companies is concerned, zaibatsu companies and non-zaibatsu companies engage in merger activity at thesamelevel. The discussion above is based on the number of bidder companies that engaged in merger transactions. It only considered whether the company bought, or merged with, another company. The analysis did not examine the number of merger transactions of the companies. It is possible that zaibatsu companies merged with more target companies than non-zaibatsu companies, herewith initiating more M&A transactions. We look at the

12 For companies that do not have widely held shares.

28 Section 2-1 Pre-war mergers

average number of merger transactions that bidder companies engaged in. Bidder companies that were part of a zaibatsu group initiated 2.2 transactions and non-zaibatsu bidder companies 2.4 transactions. We find that four zaibatsu companies and 29 non- zaibatsu companies were engaged in five or more merger transactions. Based on our data, we do not find evidence that zaibatsu companies were more active in merger activity than non-zaibatsu companies in the pre-war period. This holds for the number of zaibatsu companies engaged in mergers, as well as the number of merger transactions the zaibatsu companies initiated.

Table 2-2 Mergers by size of bidder and target in the period 1906-1937 Bidder Target All mergers <5mlnyen 5-10mlnyen 10-50mlnyen >50mlnyen < 1 mln yen 34 75 50 23 8 1-5 mln yen 29 23 45 37 4 5-10 mln yen 14 0 3 24 16 10-50 mln yen 18 2 3 17 49 >50 mln yen 5 0 0 0 22 Total 100% 100% 100% 100% 100% Notes The sample consists of 537 non-financial listed bidder companies that engaged in absorption mergers, consolidations and tender offers in the period 1906-37. The target size is measured as the written-up capitalization of the bidder company’s equity capital at the time of the merger. The size of the bidder companies is the equity capitalization as of 1937. Source Kabushiki Gaisha Nenkan (Joint-stock company Yearbook) of 1937. The corporate history of 537 non-financial companies listed on the TSE in the year 1937 was analyzed on M&A activity.

Table 2-2 provides information of merger transactions by the size of the target and bidder companies. The table shows the percentages by size-class of target companies for the size- class of bidder companies. The column of all merger transactions shows that most target companies belong to the size group below 5 million yen (63% of the total). Target companies in the size groups 5-10 million yen, 10-50 million yen and over 50 million yen are 14%, 18% and 5% respectively. Except for bidder companies with an equity capital over 50 million yen, most target companies in mergers belong to the size group below 5 million yen. Regarding bidder companies in the group of 10-50 million yen, the percentage falls to 60%, and target companies with equity capital in the same group form 17%. Bidder companies with equity capital over 50 million yen primarily engage in mergers with target companies at a value of more than 10 million yen, accounting for 71% of the target companies in this group (sum of 49% and 22%). We conclude that the target companies in the pre-war period were small, both in absolute size and relative to the bidder company. We also see that, except for the largest bidder companies (over 50 million yen), all bidder companies primarily merged with the smallest target companies (lower that 5 million yen).

Industrial sectors - We can distinguish 4 sub-periods in our analysis of the period 1906- 37. Figure 2-1 shows that the periods 1920-23 and 1926-28 are characterized by concentrated merger activity. In the periods 1916-19 and 1929-33, on the other hand, both the amount and number of mergers is significantly lower. Merger activity increases slightly in the period 1934-36. The two periods of high merger concentration in the 1920s are dominated by the electricity sector. In 1920-23 the electricity sector accounted for 35% of all merger cases, followed by textile and other manufacturing at respectively 19% and 13%. When measuring by transaction amount of the mergers, the share of the electricity sector is 59%.

29 Chapter 2 M&A activity in Japan

The amount of the mergers in the manufacturing sector is 12%, which indicates that the target companies in the manufacturing sector are larger than those in the textile sector. In the period 1926-28 the transport sector replaces the textile sector in the top 3 by number of mergers. The mergers are still concentrated in the electricity sector with 68% of the total amount and 52% of the total number of mergers. The most important reason for the high number of mergers in the electricity sector was the modernization of the Japanese economy in the 1920s. Nakamura (1994) indicates that mechanization and electrification increased rapidly in factories and even the smallest companies in the 1920s. Herewith the electrification ratio rose rapidly and the electricity companies, in order to increase efficiency and create economies of scale, initiated mergers.

Table 2-3 Mergers by industrial sector in the period 1916-1933 Number of mergers Transaction size 1916-19 1920-23 1926-28 1929-33 1916-19 1920-23 1926-28 1929-33 Electricity and gass 25 35 52 30 28 59 68 39 Other manufacturing 25 13 10 21 26 12 11 32 Textile 23 19 10 13 24 9 2 6 Mining 6200 7800 Cement 6233 2121 Chemical 6254 2206 Land transport 4 5 15 21 5 9 9 11 Steel 4101 6002 Landandwarehousing2120 0010 Shipping and shipbuilding 0 0 2 1 0 0 5 0 Other 0124 0004 Total 100% 100% 100% 100% 100% 100% 100% 100% Notes The sample consists of 537 non-financial listed bidder companies that engaged in absorption mergers, consolidations and tender offers in the period 1906-37. The transaction size of the merger is measured as the written-up capitalization of the bidder company’s equity capital at the time of the merger. The percentages are calculated as the average number of mergers and transaction amount by industrial sector for the respective sub-period. Source Kabushiki Gaisha Nenkan (Joint-stock company Yearbook) of 1937. The corporate history of 537 non-financial companies listed on the TSE in the year 1937 was analyzed on M&A activities.

Regarding the period 1916-19, table 2-3 shows that the shares for the industrial sectors textile, electricity, and manufacturing are similar with reference to the number of merger transactions and total amount. In the period 1929-33 the amount and number of mergers is largest for the electricity sector, but only at respectively 39% and 30% of the total. Transport and manufacturing both account for 21% of the number of mergers, and 11% and 32% of the total amount respectively. In 1934-36, when merger activity increased, most activity takes place in the manufacturing sector. Miyajima, Imajoh and Kawamoto (2007) find similar results from their sample of 200 companies, but they argue that three merger waves occurred in pre-war Japan: the first wave in 1900-13, the second during the 1920s, and the third in the 1930s. As the number of mergers is low in the entire sample period, not exceeding an annual total of 41 merger cases, and activity is concentrated in only one industrial sector, we cannot term these periods as a ‘merger wave’, according to our own definition. Our data only indicate that mergers between companies increased during the years 1920-23, 1926-28 and 1934.

30 Section 2-2 Post-war mergers

2-2 Post-war mergers R In the post-war period we identify two merger waves: the first in the 1960s and the second in the 1990s. R The first merger wave in the 1960s is characterized by mergers to promote industrial growth in the sectors Construction, Chemicals, and Iron and steel. Other sectors were reorganized by mergers, such as Retail and wholesale, Transportation, and Services. R The largest mergers in the 1990s occurred in the sectors Electrical equipments, Non- ferrous, Chemical, and Paper.

In this section we examine the developments in merger activity after WWII. We analyze this period chronologically, combining our own data on mergers in Japan with discussions in the literature. Our own data is derived from two sources; (i) the annual reports from the Japan Fair Trade Commission (“JFTC”), and (ii) the information from the company Recof. The JFTC annual reports provide a wealth of information on domestic merger activity, because companies were required to report their mergers according to the Antimonopoly Act up to 1997.13 JFTC therefore provides all information on merger activity in Japan, whereas the company Recof provides only publishes publicly available information on M&A activity. Below we show that, if thoroughly examined and not limited to the aggregate data in the summary, the JFTC data present interesting information. With the data we examine the number of targets and the targets’ amount of assets, and also look into the data by industrial sector when available. In some cases we add the average size of the target companies by industry, which indicates the relative importance of the industry’s mergers in a merger wave. In order to examine whether the mergers are dominated by large bidder companies that merge with small target companies, we look into the structure of the mergers based on the total equity value or total assets prior to the merger, if the data permit.

13 The JFTC data is often criticized and not used. For example, Milhaupt and West (2001) write about the data for the period 1969 to 1989 that they “may significantly overstate the level of merger activity in Japan. The JFTC data include mergers of tiny firms (including limited liability companies and partnerships) and intra-group mergers.” (p.13) Based on their data for the period 1969 to 1989, they conclude that “Japan has not experienced significant merger “waves” as in the United States.” (p.13)

31 Chapter 2 M&A activity in Japan

3,000 35,000,000 Second merger wave

30,000,000 2,500

25,000,000

2,000 yen mln in Amount

20,000,000 First merger wave 1,500 15,000,000 Number of cases 1,000 10,000,000

500 5,000,000

0 0

4 0 2 8 0 6 4 0 2 50 52 58 66 72 9 9 74 80 82 88 96 1 1 195 1956 19 196 196 1964 19 196 197 19 19 197 1978 19 19 198 1986 19 199 199 1994 19 Merger cases Target's assets

Figure 2-2 Mergers in the period 1950-1997 Notes The graph shows the number and transaction amount of all domestic mergers of non-financial companies. The transaction amount is measured as the target companies’ total assets prior to the merger. Sources Number of cases from various issues of JFTC. Transaction amount from the JFTC annual report in each year in the sample (this information is available for the years 1972-74 and 1978-1997).

Figure 2-2 shows the number of merger cases from 1950 to 1997, and the amount of total assets for the years in which this information is available (1972-74 and 1978-97). We can distinguish four sub-periods in merger activity in the post-war period: 1) 1950s: reconstruction and reorganization mergers (zaibatsu revival mergers). 2) 1963-72: First Japanese merger wave; the number of merger cases increased considerably compared to the preceding period. 3) 1985-89: increasing trend in M&A activity, primarily by Japanese companies in foreign countries. We have not termed this as a merger wave, because the size of the target companies in domestic mergers remained small and similar to the preceding period. 4) 1991- present: Second Japanese merger wave; the number of merger cases is high, and the transaction size of M&A activity shows a strong increase.

The pre-war zaibatsu groups were broken up after the war and the related companies were prohibited to use the zaibatsu company names, logos and symbols. These measures were implemented and largely adhered to until the friendship treaty was signed in 1952 and the occupation forces transferred control to the Japanese government.14 Alotofcompanies that had been broken up re-established their pre-war relations and connections in the 1950s. In the Mitsubishi zaibatsu, for example, dissolution targets were the companies Mitsubishi Trading, Mitsubishi Heavy Industries, Mitsubishi Mining and Mitsubishi Kasei. Mitsubishi

14 These measures are explained in detail in sub-section 3-2-2.

32 Section 2-2 Post-war mergers

Trading was broken up in more than a hundred companies, Mitsubishi Heavy Industries in East Japan Heavy Industries, Central Japan Heavy Industries, and West Japan Heavy Industries, Mitsubishi Mining in Mitsubishi Mining and Taihei Mining, and Mitsubishi Kasei in Nihon Kasei Industries, Asahi Glass and Shinko Rayon (Johnson 1982; Hoshino 1984; Hadley 1970). The strategic reorganization of Mitsubishi Trading started in the years from 1949 to 1951 when smaller companies combined their operations and re-established their pre-war relations. After 1952 larger trading companies merged and the pre-war company Mitsubishi Trading largely revived. Mitsubishi Heavy Industries merged the operations of the three separated companies in 1964. Overall, mergers between reorganized companies in order to return to the pre-war structure were common in all former zaibatsu groups during the 1950s and 1960s (Johnson 1982; Hoshino 1984; Hadley 1970; Miyajima 2007).

2-2-1 First merger wave The first merger wave in Japan occurred in the period 1963-72. The strong increase in the number of mergers was the most important difference between the wave and the preceding period 1953-62. The annual average of cases was 1,031 during the wave, considerably higher than the average of 431 cases in the preceding period. In contrast to merger waves in the U.S., that increase and decrease, waves in Japan increase and remain at the higher level. 15 After our examination of the size of the merging companies we turn to the industrial sectors in which these took place.

Size - Table 2-4 shows the post-merger equity capital of the new company after the merger. It presents the average number of merger cases by equity size for the period 1963-72, the preceding period 1953-62, and the subsequent period 1973-76. The period 1953-62 is dominated by mergers of small companies with 57% of all cases at an equity value of less than 10 million yen. Mergers at a value between 10 and 50 million yen and over 50 million are respectively at 25% and 18% of the total in the period before the wave.

Table 2-4 Mergers by size in sub-periods of 1953-1976 1953-62 1963-72 1973-76 Post-merger equity No. % No. % No. % <10mlnyen 247 57 369 36 235 24 10 - 50 million 108 25 393 38 426 43 50 - 100 million 30 7 100 10 118 12 100 - 500 million 32 7 115 11 142 14 0.5 - 1 billion 6 1 19 2 25 3 1-5billion 9 2 25 2 24 2 5 - 10 billion 2 0 4 0 3 0 10 - 50 billion 2 0 7 1 9 1 Total 433 100% 1,031 100% 980 100% Notes Size is measured as equity capital from the new combined after the merger for non-financial companies. The number of mergers is the annual average for the respective sub-period. The percentages show the share of the mergers in each period by equity size. Source Various issues of JFTC annual report.

The size of the mergers increases considerably during the merger wave. The percentage of the smallest mergers falls by 21 percentage-points to 36%, while the percentage of mergers

15 Please refer to sub-section 8-3-2 for a discussion and presentation in graphs of the five U.S. merger waves.

33 Chapter 2 M&A activity in Japan

at a value 10-50 million yen increases to 38% (up by 13%). The number of mergers at a value of more than 50 million yen increases to 26% (up by 8%). During the merger wave, mergers were predominantly in the equity class 10-500 million yen at 59%, a percentage similar to the mergers lower than 10 million yen in the period 1953-62. In the period following the merger wave, from 1973 to 1976, the percentage of small companies continues to fall and the percentage of large companies to increase. In the period 1973-76 the mergers with a post-merger equity capital of more than 100 million yen account for 20% of the total sample.

Table 2-5 Mergers by industrial sector in sub-periods of 1953-1976 1953-62 1963-72 1973-76 % Top5 % Top5 % Top5 Retail and wholsale 27.3 1 30.3 1 32.6 1 Machinery 10.8 2 8.2 4 7.3 Transportation and warehousing 9.2 3 9.2 2 7.3 5 Food 8.3 4 5.1 2.8 Chemical 5.9 5 3.2 2.8 Textile 5.8 4.3 3.2 Miscellaneous services 5.3 8.6 3 10.1 2 Construction 4.8 5.2 8.5 4 Iron and steel 4.1 5.2 4.1 Real estate 3.8 7.5 5 9.2 3 Wood 2.7 2.3 2.2 Other manufacturing 2.1 1.6 1.0 Printing 2.0 1.7 1.2 Glass, cement and ceramics 1.6 1.9 2.1 Other 1.5 0.9 0.4 Mining 1.5 1.0 1.1 Agriculture and fishery 1.0 1.1 1.0 Paper and pulp 0.9 1.0 1.2 Rubber products 0.7 0.7 0.5 Electricity and gass 0.3 0.2 0.2 Total 100% 100% 100% Notes The number of mergers is the annual average for the respective sub-period. The percentages show the share of mergers in industrial sectors by sub-period. Industrial sector is classified as in the annual report of Japan Fair Trade Commission. The column “Top5” indicates the industrial sectors in which most mergers occurred. Source Various issues of JFTC of annual report.

Industrial sector - Table 2-5 shows the proportion of mergers that occurred in the 20 industrial sectors during the merger wave, the preceding and following period. Although not indicated in the table, each industrial sector showed an increase in merger activity between the periods 1953-62 and 1963-72, which confirms that the latter period can be termed as a merger wave. During the wave most mergers occurred in the Retail and Wholesale sector with 30.3% of all mergers, followed by Transportation (9.2%), Services (8.6%), Machinery (8.2%), and Real Estate (7.5%). The table shows that in the periods preceding and following the merger wave most mergers occur in the Retail and Wholesale sector as well. The reason for the high number of mergers in this sector lies in the Japanese distribution system, also referred to as distribution keiretsu. Companies that have a downstream relationship in the production and distribution process form a distribution keiretsu. The multi-layered Japanese distribution system has the following features: (i) retail sector with large number of very small shops, and (ii) wholesale sector with stratified structure and segmentation by product type. These

34 Section 2-2 Post-war mergers

characteristics result in highly segmented sales channels built for each product group. Japanese wholesalers do not supply shops with 70% or 80% of the product assortment, but specialize in one specific category of merchandise. In order to build their own marketing channels, product manufacturers began selecting and setting up wholesalers throughout Japan as sales agents. Especially the development of supermarkets in the 1960s resulted in rationalization and reorganization of distribution operations, which resulted in a high number of mergers in this sector (Kawabe 1989; Suzuki 2002).

The technological innovations that took place in the corporate sector from the 1960s onward (economic take off of Japan), were part of a general modernization wave. In 1964, Japan announced its conformance with article 8 of the IMF charter, which stipulates the elimination of controls on capital transactions16, and this further spurred modernization and high growth. Japan’s trade balance began to show a surplus starting in the second half of the 1960s, although the oil crises in 1973 and 1979 caused temporary balance of trade deficits. The changing circumstances made Japanese companies realize that reorganization was necessary to compete with foreign companies. This led to various process innovations in parts production and rationalization spread from prime contractors to first-, second- and third-tier subcontractors. As a result of these developments we see a concentration of mergers in industries that depend on foreign demand and trading, such as Machinery, Transportation and warehousing, and Miscellaneous services. Other important sectors were those related to the promotion of economic growth or fundamental change in the industrial structure (new industry or new technology). Important mergers to promote industrial growth occurred in the sectors Construction, Chemicals, and Iron and steel (Odagiri 1992; Johnson 1982; Hoshino 1984; Hadley 1970). Another argument for the first merger wave is that it resulted from the fear of takeovers by foreign companies. According to Miyajima (2007), mergers in the steel and automobile industries have been the response of the capital liberalization. We revisit the first merger wave in chapter 9, where we indicate that we think the economic growth and technological innovations is a better explanation.

16 Please refer to section 4-1 for a more detailed explanation about the elimination of controls on capital transactions 1964.

35 Chapter 2 M&A activity in Japan

15 Number of cases

8 9 1 2 5 6 8 9 2 3 5 6 95 95 96 96 96 96 96 96 97 97 97 97 1957 1 1 1960 1 1 1963 1964 1 1 1967 1 1 1970 1971 1 1 1974 1 1 1977 1978 All Bidder & Target over 10bn yen Figure 2-3 Mergers between large companies in the period 1957-1978 Notes Mergers between non-financial target companies and bidder companies with an equity capital exceeding 1 billion yen prior to the merger. Source Annual report of JFTC for each year in the sample (retrieved from information on large mergers in these reports).

Large mergers - Figure 2-3 shows mergers of which the target company and bidder company have an equity value of over 1 billion yen. The figure covers the period 1957-78, and the trend of these large mergers is similar to the overall merger activity in figure 2-2. The peak year for large mergers was 1963; most cases occurred for all large mergers as well as those with combined equity value of over 10 billion yen. After 1963 the number of merger cases declined, and in 1966 and 1967 the mergers between companies over 10 billion yen drops below 5 merger cases. Higher merger activity occurs in the following years, but the mergers over combined equity value of over 10 billion yen are negatively affected by the oil crisis in 1973; the number of merger cases with equity value below 10 billion yen remains relatively high after this year. Regarding the mergers aimed at promotion of economic growth, the government and bureaucracy took an active role. Government policy was aimed at improving competitive strength of Japanese companies in relation to foreign companies and preventing excessive competition. In this respect, mergers were important, enforcing both policy goals. In the shipping industry this resulted in 6 horizontal mergers of 12 shipping companies in 1964, such as Mitsui Sempaku and Osaka Shosen into Mitsui O.S.K. Line, and Nippon Yusen and Mitsubishi Shipping into Nippon Yusen. The steel industry was also a targeted industry and various mergers were undertaken in order to achieve economies of scale in large-scale steel enterprises. In 1965 Kobe Steel merged with Amagaseki Seitetsu, in 1967 Fuji Iron and Steel merged with Tokai Seitetsu, and in 1970 Yawata Iron and Steel merged with Fuji Iron and Steel. MITI played an important role in the merger between Yawata Iron & Steel and Fuji Iron & Steel to form Nippon Steel in 1970. The two steel makers shared a 35.6% of crude steel production, which exceeded the JFTC criterion of 30%. The

36 Section 2-2 Post-war mergers

JFTC rejected the merger but under pressure from MITI approved after a formal hearing. MITI was also involved in the merger between the auto companies Prince Motors and the Nissan Group Corp. in 1966 (Caves and Uekusa 1976; Kester 1991; Hasegawa 1996; Hart 1992). Friedman (1988) argues, however, that this merger was not in line with MITI’s plans. Initially, MITI wanted to consolidate the industry and arrange all automotive companies behind Nissan and Toyota. This proposal was opposed by the automotive companies and MITI amended the plan to form three companies; Toyota, Nissan, and a combination of Toyo Kogyo, Fuji, and Isuzu. These plans and additional amendments were not accepted either by the automotive industry and the number of automotive companies had increased to seven by 1970 (Christelow 1995; Friedman 1988).

2-2-2 Latter half 1980s The latter half of the 1980s is characterized by high corporate profits and a low number of corporate bankruptcies. Companies made investments in plant and equipment and a lot of urban development took place. This situation led to rising stock and land prices in the late 1980s, the stock market index reaching a maximum of 38,915 in December 1989 and land prices peaking in September 1991. Japanese companies were very active in M&A activity in this period, both domestic and in targeting companies and investment projects outside Japan. Odagiri (1992) indicates that there was an increasing trend in domestic M&A activity, primarily aimed at diversifying company’s operations. The high stock prices combined with the loosening of controls on international capital controls and easy monetary policy boosted Japanese M&A activity abroad in real estate and companies. Outward real estate investment increased from 2.5 billion dollars in 1985 to 46.4 billion dollars in 1990. These investments were made in well-known commercial properties, shopping centres, golf courses, hotels, and undeveloped land in the states of Hawaii and California. For example, Mitsubishi Real Estate bought an interest in New York City’s Rockefeller Plaza at a value of 120 billion yen (1.46 billion dollars) in 1989. Real estate investments in Europe were concentrated in London with for example the purchase of Britannic House, Moor Lane by EIE International and the Old Bailey by Mitsui Real Estate. Real estate investments in the early 1980s were primarily made by construction, development and trading companies (notably Kumagai Gumi, Itoh, Shimizu and Kajima), and in the late 1980s life assurance companies were the most active investors (Mason 1997; Lapier 1998). Japanese companies were also active in acquisition of foreign companies such as the acquisition of Firestone Tyer and Lever by Bridgestone in 1988 for 333,190 billion yen, Columbia Pictures by Sony in 1989 for 644,000 billion yen, and MCA by Matsushita Electric Industrial Co., Ltd. in 1990 for 780,000 billion yen (Recof 2003).

2-2-3 Second merger wave The stock market bubble burst after its peak in 1989 and lost more than half its value by 1992. After the land price bubble burst in 1991, land prices showed a continuing fall in the 1990s and 2000s. With the burst of the two bubbles, many indebted companies had difficulty in making interest payments and repaying debt. The debt was mortgaged by land and real estate, and resulted in non-performing loans for financial institutions during the 1990s. The bursting of the bubbles and the subsequent economic recession led a lot of companies to initiate M&A activity as a means to restructure operations. The second merger wave coincides with these developments and starts in 1991. The wave is still

37 Chapter 2 M&A activity in Japan

continuing at present but can be divided into two periods; the first period covers the years 1991-97 and the second period starts in 1999 and continues at least until 2006 (our data-set ends in December 2006). Table 2-6 provides details on the mergers in the two periods of the second wave with data from JFTC (up to 1997) and Recof. The table gives the average of merger cases of both sources, the amount of domestic mergers from JFTC and the amount of all M&A activity from Recof.

Table 2-6 Merger activity by number of cases and by amount in the period 1984-2006 JFTC data Recof data Target assets in Merger M&A amount Merger Cases M&A cases mln yen Cases in mln yen 1984-1990 1,453 2,728,963 48 517 -- 1991-1997 2,487 22,328,845 204 772 -- 1999-2006 -- -- 625 2,862 11,088,627 Notes JFTC data: average number of all non-financial domestic mergers and annual average of targets’ total assets for the sub- periods. Recof data: all M&A activity involving a Japanese company for which information was publicly available. Sources Annual report of JFTC for each year in the sample. Recof (2003, 2007).

The JFTC and Recof data show that the number of mergers in the period 1991-97 increased considerably compared to the period 1984-90, respectively by 71% and by more than 3-fold. In the period 1999-2006 the Recof data show an average of mergers at 625 cases and an average of 2,862 cases for total M&A activity. In this period the number of M&A cases increased by 2,090 cases, compared to an increase of 255 between the periods 1984-90 and 1991-97. Table 2-6 clearly indicates that the size of the mergers was the most important characteristic of the second wave. The JFTC data show that, whereas the number of merger cases increased by 71% between the two periods, the size of the target companies, measured as the target company’s total assets, increased more than 8-fold from 2,700 billion yen in 1984-90 to 22,300 billion yen in 1991-97. The amount provided by Recof is not comparable with the JFTC data, as it reveals only the amount of M&A cases for which information was publicly available. JFTC shows the target company’s assets size of all domestic mergers. Compared with the first merger wave we have more JFTC information available for the first period of the second merger wave. This allows us to do a thorough analysis and we start by looking at industrial sector level with the number of targets, the total size of the targets, and the average size of the targets. In the analysis we compare the period 1991-97 with the preceding 7-year period. Table 2-7 indicates that we can refer to the increased merger activity in 1991-97 as a wave because all 24 industrial sectors we investigate have an increase in mergers measured by number, size and average, compared to the preceding 7-year period: Whereas in 23 industries the value measured as total assets and average assets more than doubled during the wave, the number of targets more than doubled in only 5 industrial sectors. We can conclude that the size of the targets was considerably larger during the wave.

38 Section 2-2 Post-war mergers

Table 2-7 Comparison of merger activity by industrial sector between 1984-1990 and 1991-1997 No. of targets Total assets Average assets Increase >100% 5 23 23 50-100% 14 0 0 0-50% 5 1 1 Decrease 0-50% 0 0 0 50-100% 0 0 0 Notes Merger cases are compared based on annual averages for the non-financial industrial sectors in the periods 1984-90 and 1991-97. The number of mergers is measured as the number of target companies, total assets as the target companies’ total assets prior to the merger. Average assets is calculated by dividing targets’ total assets by the number of targets in an industrial sector. Source Annual report of JFTC for each year in the sample.

Similar to the first wave, most mergers occurred in the sector Wholesale and retail with almost 30% of all cases and 32% of the total transaction amount. The sectors Real estate and Miscellaneous services were second and third with percentages of respectively 11% and 10% of the total transaction amount. The table shows the averages for individual mergers in each industrial sector as well, and reveals that the three sectors Wholesale and retail, Real estate and Miscellaneous services do not belong to the top three according to this measure. In fact, the sectors’ percentages do not exceed 3%. This implies that the merger activity was dominated by a lot of mergers involving small target companies in these industries. The average size of a merger is highest in the industries Electrical equipments, Non-ferrous, Chemical, and Paper and pulp. The industrial restructuring was the result of mergers by large companies during the period 1991-97. We examine the influence of mergers between large companies on the overall structure within an industrial sector. We measure the importance of merger activity in a sector by dividing the total asset amount of target companies by total assets of all companies in the respective industrial sector. 17 By this measure large mergers are concentrated in the industries Paper and pulp, and Chemical.18 The mergers had an important effect on the structure of the sectors with a ratio of target companies’ total assets involved in mergers at 31.8% in the Paper and pulp industry, and at 39% in the Chemical industry. In 1993 the company Jujo Paper merged with Sanyo-Kokusai Pulp to form Nippon Paper industries, and Oji Paper merged with Kanzaki Paper to form New Oji Paper. New Oji Paper merged with Honshu paper in 1996 and the new company was renamed to Oji Paper. The first large merger in the chemical industry occurred in 1994 when Mitsubishi Chemical Corporation merged with Mitsubishi Kasei Corporation to form Mitsubishi Petrochemical. A year later Taiyo Toyo Sanso merged with Taiyo Sanso into the company Toyo Sanso. In 1997 Mitsui Chemicals was formed when Mitsui Petrochemical Industry merged with Mitsui Toatsu Chemicals, Inc.

17 Total assets of target companies calculated form data yearly issues of the JFTC annual reports. Total assets of all companies in industrial sectors retrieved from Ministry of Finance Japan, Policy research institute “http://www.mof.go.jp/1c002.htm” 18 Also in the banking sector, but we focus in our discussion on non-financial companies.

39 Chapter 2 M&A activity in Japan

Table 2-8 Mergers by industrial sector in the period 1991-1997 Total assets No. of targets Average assets %Top5%Top5%Top5 Retail and wholsale 29.7 1 32.0 1 3.0 Real estate 10.7 2 11.1 3 2.1 Miscellaneous services 9.5 3 16.1 2 1.6 Chemical 8.5 4 2.4 9.9 3 Electrical equipments 8.0 5 2.4 13.6 1 Transportation and warehousing 4.6 4.6 5 4.0 Machinery 4.2 1.9 5.0 Construction 3.6 6.8 4 1.3 Metal 3.0 1.4 7.2 5 Non-ferrous metals 2.5 0.4 12.4 2 Textile 2.5 1.9 3.9 Food 2.5 2.3 4.0 Paper and pulp 1.9 0.5 7.4 4 Transportation equipments 1.7 0.9 3.6 Glass, cement and ceramics 1.4 1.5 2.0 Other manufacturing 1.3 1.6 1.9 Printing 1.1 1.5 2.5 Iron and steel 1.0 0.6 4.4 Precision instruments 0.8 0.6 3.9 Agriculture and fishery 0.3 0.4 2.4 Wood 0.3 0.7 1.0 Rubber products 0.2 0.2 1.9 Mining 0.1 0.1 0.5 Electricity and gass 0.0 0.1 0.5 Total 100% 100% 100% Notes Percentage of mergers by non-financial industrial sector for the number of targets, target’s total assets and targets’ average assets. The number of mergers is measured as the number of target companies, total assets as total assets of the target companies prior to the merger. Average assets is calculated by dividing the targets’ total assets in an industrial sector by the number of targets in that sector. Source Annual report of JFTC for each year in the sample.

Table 2-9 shows the number of merger cases, their amount and averages by merger type for the periods 1977-83, 1984-90 and 1991-97. The table indicates an increasing trend in each measurement of merger activity. The merger wave is clearly driven by horizontal mergers with an annual average total of 805 merger cases at a value of 8,453 billion yen. Companies faced a lot of difficulties, or even financial distress, and coped with the problems by strengthening their position in the market. Compared to the preceding period, large companies engaged in mergers affecting the structure of entire industrial sectors. The average asset value of targets is 354 billion yen which is slightly lower than the forward and backward vertical merger at respectively 426 billion yen and 485 billion yen. This indicates that the average vertical merger deal was (slightly) larger than horizontal mergers. The table also shows that vertical mergers, whether forward or backward, are similar in number and size during the period 1991-97. The importance of vertical mergers indicates that companies restructured the production process. As the forward and backward mergers were at the same level, the parent companies were engaged in rationalizing both their production channel and their sales channels. Conglomerate mergers occur more frequently than vertical mergers in the three sub-periods, but the size of these mergers is considerably smaller. The average size of a transaction involving geographic expansion is 69 billion yen and the average size of a transaction involving product extension is 256 billion yen.

40 Section 2-2 Post-war mergers

Overall, the mergers in the period 1991-97 were dominated by horizontal and vertical mergers.

Table 2-9 Mergers by merger type in the period 1977-1983 Vertical Conglomerate Horizontal Geographic Product Forward Backward Other Expansion Extension No. of targets 1977-83 241 77 80 208 117 331 1984-90 439 109 135 244 172 411 1991-97 805 186 186 311 264 686 Total assets (mln yen) 1977-83 417,334 267,200 248,755 176,930 139,432 250,304 1984-90 1,893,433 276,599 1,398,132 620,996 592,463 848,982 1991-97 8,452,568 2,652,868 2,950,283 1,635,105 2,339,139 4,278,105 Average assets (mln yen) 1977-83 48,992 179,362 67,221 8,810 26,120 19,804 1984-90 75,801 39,050 111,182 65,826 95,795 51,617 1991-97 354,220 426,424 484,801 69,153 256,418 221,944 Notes The number of mergers is measured as the number of target companies, total assets as total assets of the target companies prior to the merger. Average assets is calculated by dividing the targets’ total assets by number of targets for each merger type in an industrial sector. Mergers are in non-financial industrial sectors. Source Annual report of JFTC for each year in the sample.

Second period The second period of the second wave is characterized by the many changes in the laws and regulations regarding M&A. The Antimonopoly Act was amended in 1997 and created the possibility to set up holding companies19 and in this year simplified merger procedures were also introduced. Since 1999 the Japanese Commercial Code permits Japanese companies to use stock swaps and stock transfers for corporate acquisitions and reorganizations. As of fiscal year 1999 Japanese companies had to adopt new accounting standards, requiring companies to disclose consolidated rather than parent-only earnings in financial reports and to report financial assets at market value instead of book value. The corporate reorganization tax reforms have been effective since 2001 and allow for easier corporate divestures. For our analysis of the second period of the second wave we turn to Recof data. As already discussed, this data is different from the JFTC data as it looks into all M&A activity for which information is publicly available, in stead of only, but all, domestic mergers. Further, the amount is not the amount of all transactions but only of those transactions for which the amount was available. The data allows us to look more closely at the difference in M&A activity between the two periods of the second wave. We focus on the kind of M&A activity and whether companies that engage in the activities belong to the same horizontal corporate group.

19 Establishment of a pure holding company had been prohibited since 1945. Please refer to sections 3-2 and 3-3 for a detailed discussion.

41 Chapter 2 M&A activity in Japan

4,000 Second wave - Period 1 Second wave - Period 2

3,500

3,000

2,500

2,000

Number of cases 1,500

1,000

500

998 002 003 005 006 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1 1999 2000 2001 2 2 2004 2 2

Merger Acquisition Business Transfer Capital Participation Increased Investments Figure 2-4 M&A cases in the period 1985-2006 Source Recof (2003, 2007)

Figure 2-4 shows that total M&A activity increased considerably in the latter half of the 1990s. Until 1996 total M&A activity did not exceed 700 cases, it increased to 1,227 in the year 1998 and jumped to 1,813 in the year 1999 to reach 3,755 cases in 2006. In the years 1997 and 1998 the number of merger cases was 357 and 358 respectively, showing only a small increase compared to 1996 with 318 cases. After stock swaps and stock transfers were permitted in 1999, the number of merger cases increased with more than 200 transactions to 569 cases in 1999. Interestingly, the number of merger cases remained stable in the range of 600 cases during the period up to 2006. Acquisitions, resulting in ownership exceeding 50%, also showed a strong increase in 1999 with 124 transactions to 435 cases. The number of acquisitions continued to rise with the implementation of the tax-free restructuring law in 2001 and Industrial Revitalization Law in 2003, reaching totals of 1,211 and 1,246 cases in 2005 and 2006. Business transfers totaled 623 cases in 2001, showing a strong increase of 138 transactions as the tax-free restructuring law allowed easier business divestures and sell-offs of assets as part of corporate restructuring. Business transfers increased to 700 cases in 2002 and this level has remained stable. Capital participation increased sharply to 725 cases in 2000, dropped the following 3 years to 487 to increase again thereafter. Increased investment shows a smooth upward trend.

42 Section 2-2 Post-war mergers

Table 2-10 M&A activity categorized by in-group and out-group cases in the period 1990-2006 1990-1997 1998-2006 In-group Out-group In-group Out-group Merger 76.3 23.7 86.0 14.0 Acquisition 7.2 92.8 11.7 88.3 Business Transfer 19.1 80.9 36.8 63.2 Capital Participation 3.3 96.7 1.9 98.1 Increased Investment 26.5 73.5 14.1 85.9 Total 25.4 74.6 31.7 68.3 Source Recof (2003, 2007)

Table 2-10 shows the percentages of the M&A activity by in-group and out-group transactions. Of all transactions, the percentage of in-group transactions increased in the period 1998-2006. This higher percentage for in-group transactions was primarily caused by the increase in in-group mergers (76.3% to 86%) and business transfers (19.1% to 36.8%). These two types of M&A were primarily used to restructure the organization of group companies as a result of the new regulations. Over the period 1990-2006, we see that acquisition, capital participation and increased investment predominantly occurred between companies that did not belong to the same group. The possibility to form a holding company and to use stock swaps and stock transfers, combined with the need for industrial restructuring resulted in large M&A deals in the financial services and telecommunications sectors. In 2000 Mizuho Holdings was formed in a merger of the Dai-Ichi Kangyo Bank, Fuji Bank and Industrial Bank of Japan. The Bank of Tokyo-Mitsubishi, Mitsubishi Trust and Banking, and Nippon Trust Bank formed the Mitsubishi Tokyo Financial Group. Tokai Bank and Sanwa Bank merged into the UFJ Bank in January 2001. In the telecommunications sector DDI, KDD, and IDO (Nippon Idou Tsushin) formed the new company, KDDI. Japanese companies in financial distress provided opportunities for foreign companies to enter the Japanese market. In 2001 Manulife Century Life Insurance acquired Daihyaku's insurance contracts following its bankruptcy, and American International Group (AIG) acquired the bankrupt insurance company Chiyoda Mutual Life. In 2002 the most important goal of the government was to clean up bad debts of companies and banks tried to revitalize important customers with debt-equity swaps. Examples are the debt-equity swap of (i) Daiei by UFJ Bank, Sumitomo Mitsui Banking Corporation, Mizuho Corporate Bank, (ii) Haseko Corporation by Daiwa Bank, Chuo Mitsui Trust and Banking, Mizuho Corporate Bank, and (iii) Orient Corporation by Mizuho Corporate Bank. Foreign investment firms were also involved in restructuring Japanese financial institutions, such as Ripplewood Holdings’ acquisition of the Long- Term Credit Bank of Japan and Lone Star’s acquisition of First Credit (bankrupt non-bank that had been affiliated with the Long-Term Credit Bank of Japan).

43 Chapter 2 M&A activity in Japan

2-3 Hostile takeovers R The first hostile takeovers in the post-war period occurred immediately after WWII and they continued during the entire period. R Greenmail is the practice of a target company buying back its own shares from a hostile bidder at a substantial premium, but can also result from an unsuccessful hostile tender offer attempt.

As we explained in chapter 1, the difference between hostile tender offers and greenmail lies in the motivation of the bidder (raider). Whereas in a hostile tender offer a bidder tries to attain operational control over the target company, in a greenmail the bidder wants the target company to buy back its own shares at a substantial premium. In post-war Japan greenmail has primarily been done by so-called shite groups that, after the purchase of a large number of shares, “threaten to use special large-shareholder rights (e.g., to inspect the company’s books or call a special shareholders’ meeting) to hamper management’s control” (Kester 1991, p.245). The two activities are closely related because an unsuccessful hostile tender offer attempt is likely to evolve into greenmail. Distinguishing an unsuccessful hostile tender offer from greenmail is therefore difficult; greenmail can be the result of buying up shares with the intention to greenmail, or the result of an unsuccessful hostile tender offer. This implies the possibility that an unsuccessful hostile tender offer is interpreted as greenmail. In line with previous research, we use the term hostile takeover for both activities when multiple interpretations are possible. In chapter 7 we revisit the cases discussed below, and explain why we think the hostile takeovers in this section need to be interpreted as greenmail. The first hostile takeovers in the post-war period occurred in the 1950s. The Korean War had increased the wealth of various businessmen and, combined with a falling stock price of many companies, hostile takeovers towards former zaibatsu companies occurred “where shares have been bought up quietly on the market” (Okumura 2000, p.65). Youwa Properties, a company that had been part of Mitsubishi real estate (a first-tier company of the Mitsubishi zaibatsu), was confronted with stock cornering in 1952. This resulted in the stock price to increase from 323 yen in January 1952 to 1,600 yen in August 1952 (Okumura 2000). In August 1952 the takeover group’s ownership of Youwa Properties had reached 34%, and a seat on the board of directors was requested (Miyajima 1994, p.318). Mitsubishi Bank, together with eight former zaibatsu companies, bought back the shares in September 1952. The zaibatsu companies responded to these hostile takeovers as a group by “coordinating white knight and white squire defensive arrangements to protect their former affiliated companies from hostile takeovers” (Morck and Nakamura 1999). Another well-known case is the hostility by Hideki Yokoi towards Shirokiya, a department store company. Mr. Yokoi had become wealthy from dealings on the black market, with the military and Supreme Commander for the Allied Powers (“SCAP”). In 1953 he had purchased more than 40% ownership of Shirokiya and won control over the board after a general stockholder’s meeting. Four days later Yokoi lost this control because of illegal trading. The chief executive of the Tokyu Store, also a department store company, mediated in the process. Morck and Nakamura (2003) argue that “[Y]okoi continued launching corporate takeovers over the following two decades. After Shirokiya, he mounted raids on Toa Oil, Daikyo Oil, Imperial Hotels, Tokai Shipping Line, Toyo Sugar, Shibaura Sugar, Taito Sugar, Dainippon Sugar and many other companies” (p.77: with

44 Section 2-3 Hostile takeovers

corrected spelling company name “Shirokiya”). Although we term these activities as hostile takeovers, we show in chapter 7 that these are better qualified as greenmail. Morck and Nakamura (1999) argue that a series of high-profile hostile takeover raids occurred in the two decades after the 1950s. We failed to find any hostile tender offers, and table 2-11 indicates some greenmail cases by shite-groups that occurred during the 1970s and 1980s. This is a very short list of the actual number of greenmail cases. Kester (1991) indicates that a list was published in 1988 that showed 120 private investors, companies, and groups engaged in greenmail or associated risk-arbitrage in 156 listed companies. Together they had positions with a market value of 1.5 trillion yen in 1988, of which 58% was in companies on the First Section of the Tokyo Stock Exchange.20

Table 2-11 Greenmail cases in the 1970s and 1980s Est. Shares Target Owner Year Purchased (%) Kao Soap Hong Kong investor group 1970s Ajinomoto Hong Kong investor group 1970s Nakayama Sasagawa 1970s Mitsui Mining Mr. Toguri 1976 ถ30 Daido Sanso Nihon Tochi 1977 Oji Paper Hong-Kong based companies 1978 13 Okamoto Sasagawa group 1978 Yuasa Battery Nihon Tochi 1980 Tokyo Nissan Shuwa 1981 22.9 Kobe Electric Railway Sasagawa group 1983 Mitsui Toatsu Nihon Tochi 1985 Toyota Automatic Loom Works Nihon Tochi 1987 7.1 Chujitsuya Shuwa 1989 33 Inegaya Shuwa 1989 25 Isetan Shuwa 1989 24 Nagasakiya Shuwa 1989 14 Sources Kester (1991); Sheard (1991)

Greenmail cases - In 1976 Mr. Toguri of the Osano group obtained approximately 30% ownership in Mitsui Mining. Ten members of the Mitsui keiretsu, coordinated by Mitsui Bank, bought the shares back and share-ownership of Mitsui Mining by group-members increased to 60% (Sheard 1989). Another stock-cornering group, Hong Kong Investor Group, bought shares in the companies Kao Soap, Ajinomoto and Oji Paper. As a Mitsui keiretsu group company, the shares of Oji Paper were repurchased by the Mitsui Bank, Mitsui Trust & Banking, other Mitsui keiretsu companies, and the Industrial Bank of Japan (Sheard 1989; Morck, Nakamura and Shivdasani 2000). Similar to the Mitsui keiretsu, repurchase of shares from stock cornering groups occurred in other keiretsu companies. Nakayama Steel faced hostilities in 1978 and, belonging to the Sanwa keiretsu, its shares were bought back by group-members and banks it had relations with (Sheard 1989, p.428). In the 1980s similar greenmail cases occurred, for example, by greenmailer Nihon Tochi in 1987. Nihon Tochi had bought 7.1% of Toyota Automatic Loom Work’s shares. These shares were repurchased by Toyota Motor, at a profit of at least 10 billion yen for Nihon Tochi (Kester 1991, p.246). Other large cases in the 1980s are the stock cornering of

20 Please refer to Kester (1989) and Sheard (1989) for more examples of greenmail cases.

45 Chapter 2 M&A activity in Japan

shares from retail companies such as Chujitsuya (33%), Inageya (25%), Isetan (24%), Matsuzakaya (16%), and Nagasakiya (14%) by a real estate company, Shuwa (Sheard 1989, p.428). Chujitsuya and Inageya, in an attempt to dilute Shuwa’s shareholdings issued a 20% equity stake in each other. Shuwa filed for a court injunction to stop the new equity issuances and the court deemed the issue unfair: there was no rational use of the funds and was only aimed at lowering a specific shareholder’s ownership in the company. Shuwa later sold its ownership in Chujitsuya to Daiei and Inageya to Aeon.21 As indicated above, various kinds of companies were targeted, including well-known companies that were part of horizontal keiretsu, such as Konica, Teijin (Sanwa group), Sapporo Breweries (Fuyo group), Mitsubishi Rayon (Mitsubishi group), and Mitsubishi Steel (Mitsubishi group) (Kester 1991, p.247).

The greenmail cases up to the late 1980s involving parties such as shite-groups, led to the first hostile takeover involving a foreign company in 1989. In April 1989 Boone Company, an investment company headed by T. Boone Pickens, purchased 20.2% of Koito Manufacturing Company from a private investor, Mr. Watanabe.22 Koito Manufacturing Company is Japan's largest maker of automotive lighting equipment and an important supplier for Toyota Motor Company that owns 19% of Koito’s shares. Mr. Watanabe, a Japanese billionaire, had been amassing Koito stock during a year and a half and was trying to have Toyota repurchase the shares. Toyota, however, after adhering to the expensive greenmail in relation to Toyota Automatic Loom Company in 1987, had sworn never to comply with such demands again. Mr. Watanabe, unable to profit from his investments, asked Mr. Pickens to act as shareholder in an attempt to put additional pressure on Koito and Toyota. When Mr. Pickens arrived in Japan he argued that he was interested in Koito as a long-term investment and asked for a seat on the board of directors. After he had been unable to get a seat on the board for 2 years he sold his shares in 1991. Whether Pickens really was a long-term investor, as he claimed himself, or was merely used as additional pressure by Mr. Watanabe, is difficult to say and opinions are divided. In our opinion the actions by Mr. Pickens are more likely due to be a greenmail attempt because of the sequence of events and the fact that he sold his stake in Koito back to Mr. Watanabe in 1991.

21 The Asahi Shimbun: February 25,2005, Livedoor sues to block radio company's tactic 22 Based on The New York Times: April 5, 1989, Pickens Group Confirms Stake; April 20, 1989, Pickens Courting a Wary Tokyo ; October 12, 1993, Hotel Sale May Signal End of Japan's Patience ; December 6, 1992, Corporate Japan's Unholy Allies. Reuters: July 6, 1991, Pickens Drops Japan Suit.

46 Section 2-4 Hostile tender offer attempts

2-4 Hostile tender offer attempts R Prior to 1999 there were no hostile tender offers in Japan and in the period from 1999 to July 2007 thirteen attempts were initiated. R Only two hostile tender offers were successful, but the “hostility” in both tender offers is doubtful as the bidder company and target company had a long-standing relationship.

In this section we look into the hostile tender offer attempts that were initiated as of 1999. As the list is short, and the cases are of great importance for the Japanese economy and our institutional analysis of hostile takeovers, we discuss all cases individually.

Table 2-12 Hostile tender offers in Japan Bidder Target Year Type Result Cable & Wireless IDC 1999 Hostile TOB Succesful Nippon Boehringer SS Pharmaceutical 2000 Hostile TOB Succesful Ingelheim Co. MAC Corp Shoei 2000 Hostile TOB Not successful Steel Partners Yushiro Chemicals 2003 Hostile TOB Dividend increase Steel Partners Sotou 2003 Hostile TOB Dividend increase Livedoor Nippon Broadcasting 2005 Off-market purchase Negotiation Yumeshin Holdings Nihon Gijitsu Kaihatsu 2005 Hostile TOB White knight Rakuten TBS 2005 Off-market purchase Negotiation MAC Shin-Nihon Musen 2005 Hostile TOB White knight Don Quijote Origin Toushu 2006 Hostile TOB White knight Oji Paper Hokuetsu Paper 2006 Hostile TOB White squire; poison pill Steel Partners Myojo Foods 2006 Hostile TOB White knight Steel Partners Sapporo Brewery 2007 Hostile TOB Poison pill Steel Partners Bull Dog Sauce 2007 Hostile TOB Poison pill Steel Partners Tenryu Saw 2007 Hostile TOB Not successful Sources Newspaper articles as indicated in footnotes 23-35 of this chapter.

Cable & Wireless23 - Cable & Wireless P.L.C. (“C&W”), a British telecommunications company, initiated the first hostile tender offer bid in Japan. The target company was International Digital Communications Inc.(“IDC”),asmallphonecompanywitha minor part of the international call market. It should be noted that the use of “hostile” needs to be interpreted with care in this case; C&W was among the original founding shareholders of IDC in 1986 and can therefore be considered as an insider. The reason for the tender offer was that private negotiations to strengthen relations were not fruitful. C&W’s initial bid was in March 1999 for more than 50% of IDC’s shares at a bidding price of 100,000 yen. Although the specific amount was not disclosed, Nippon Telegraph and Telephone Corporation (“NTT”) offered a price that was a few thousand yen higher. After a board meeting in April, the IDC board rejected the offer of C&W in favour of NTT. On May 7 C&W increased its bid to 107,372 yen for 82.3% of the shares in order to obtain full ownership of the company. NTT raised its bid to 108,974 yen and C&W responded by increasing its bidding price to 110,577 yen a share. The two second largest shareholders,

23 Based on The New York Times: September 23, 1987, Japan Phone Accord Fails ; April 15, 1999, N.T.T. Says It Has Edge In Acquiring Rival Concern ; May 7, 1999, Cable and Wireless In Bid for I.D.C. of Japan ;June2, 1999, Bidding war intensifies ; June 10, 1999, Cable and Wireless Wins Fight for Japan Phone Carrier ; June 17, 1999, Cable and Wireless in Talks With Local Japanese Phone Carriers.

47 Chapter 2 M&A activity in Japan

Toyota and Itochu said they would accept the offer of C&W at this price and C&W obtained 97.69% ownership of IDC through the tender.

Boehringer Ingelheim24 - The second hostile tender offer was by Boehringer Ingelheim GmbH (“BI”), a large German pharmaceuticals company that wanted to take advantage of the deregulation of the Japanese pharmaceutical market. The target company, SS Pharmaceuticals Company (“SSP”), is a Japanese drug maker that produces over-the- counter and prescription drugs, vitamin supplements and tonics, and has a vast distribution network and its own chain of stores. BI had a long relationship with SSP, owned 19.61% in the company and, as the largest shareholder, had a seat on the 16-member board of directors. BI started its tender offer on 17 January 2000 and aimed to attain at least a 33.4% stake in SSP, a percentage that would enable the company to veto management decisions, in order to assure long-term security of the investment. BI offered a premium of 42% over the closing price the day before it made its offer, sent shareholders letters explaining the offer, and set up a toll-free number for shareholders who wanted more information. Through the tender offer BI could raise its ownership in SSP to 35.86%.

MAC Corporation25 - The third hostile tender offer attempt was the first hostile takeover involving a Japanese bidder and Japanese target. The Japanese bidder company was MAC Corp.26, an acquisition fund established by a former bureaucrat, Mr. Yoshiaki Murakami. The target company was Shoei Company, a cash-rich former silk company that had diversified into electronics and property and belonged to the Fuyo keiretsu group; Fuyo group members owned 60% of Shoei. MAC made a hostile tender offer for all shares of Shoei with an offer period from 24 January 2000 to 14 February 2000 at an offer price at 1,000 yen per share, which was a premium of 14% compared to the day before the offer. After the announcement of the bid Shoei’s share price rose considerably, reaching 1,480 yen at one point. MAC managed to increase its ownership from 2.78% to only 6.52%. It was argued that the shareholdings within the Fuyo group prevented the hostile takeover, but another explanation might be the very low premium, especially in relation to its shareholdings in Canon (Shoei spun off the company Canon years ago).27

24 Based on The New York Times: January 18, 2000, Bid for Japanese Drug Maker Is Latest Sign of Deregulation ; February 17, 2000, In Hostile Bid, Germans Buy 35% of Japan Drugmaker. 25 Based on The New York Times: January 18, 2000, Bid for Japanese Drug Maker Is Latest Sign of Deregulation ; January 25, 2000, A hostile Japanese offer; February 15, 2000, Hostile Takeover Bid in Japan Ends on an Educational Note ; March 29, 2000, Japanese Shareholder Activist Loses 2nd Battle With Company. 26 MAC is an affiliate of M&A Consulting Inc., a company that engages in advisory work and fund management. 27 Bremner (2005) explains that after this failure Murakami changed his strategy from investing millions and turning around a company to focusing on small and mid-size cash-rich companies. For example, he bought up 11% of the Tokyo Style, a small clothing company with a lot of cash and argued that the company had to share the company’s wealth with the shareholders by increasing dividend and repurchasing one third of the outstanding shares.

48 Section 2-4 Hostile tender offer attempts

Steel Partners28 - The U.S. hedge fund Steel Partners of the United States29 (“SP”) has been the most aggressive foreign raider in Japan, undertaking hostile takeover attempts since 2003. All its offers have failed, either by anti-takeover measures or friendly tender offers by other companies. SP engaged in its first hostile tender offer attempts to gain complete ownership of Japanese companies on 19 December 2003. The targets were Yushiro Chemical Industry, the largest Japanese producer of machinery lubricants, and textile dyer Sotoh Company. The management of both companies strongly opposed to the bid because it was perceived as an asset stripping takeover rather than restructuring. The tender offers failed in both cases after the two companies significantly increased dividend payouts. Being the top shareholder with 23.1%, SP targeted Myojo foods, an instant-noodle producer, with a tender offer for all its shares from October 27 to November 27, 2006. The management did not agree with the bid and found a white knight in the industry-leader Nisshin food. Nisshin made a tender offer for 33.4% of Myojo Foods’ shares at a price of 870 yen per share for the stake, exceeding the price of 700 yen per share offered by SP. SP supported the Nisshin's offer and sold its shares in Myojo Foods and made a 73% return on its failed takeover bid for Myojo Foods. In February 2007 SP targeted Sapporo Brewery, Japan's third-largest brewer, which produces Yebisu beer and distributes Guinness stout and Beringer wines in Japan, and owns real estate in Yebisu. SP had become the largest shareholder of Sapporo with 18.64% and launched a tender offer for all shares. After SP had indicated it wanted to increase its shareholdings in Sapporo, the management implemented takeover defenses after gaining shareholder approval in March. The poison pill and the so-called “advance warning system” were implemented. The latter delays a takeover offer for months or longer by requiring bidders seeking more than 20% of shares to go through a questioning process with management in which they are supposed to solicit approval for the sale. Bull-Dog is a sauce manufacturer well-known for its Worcestershire sauce. SP, owning 10.15% of Bull-Dog Sauce, declared it wanted to improve the company’s performance and made a tender offer for the remaining shares on May 16. The price with a 20% premium was set at 1,584 yen per share and raised on June 15 to 1,700 yen. The management of the company did not support the bid and gained shareholder approval on June 24 to install a poison pill defence. This defence allows Bull Dog to issue three warrants per share. SP’s request to prevent the issue was rejected by the Tokyo District Court. On May 24 SP made a tender offer for all the shares of Tenryu Saw Manufacturing.Intheofferperiodfrom May 24 to July 4 it got tenders for 2.57% of common stock, herewith increasing its ownership to only 11.18%.

28 Based on The New York Times: February 25, 2004, Sotoh Bid Fails ; February 17, 2004, Textile Concern Fights Takeover. Forbes: March 29, 2007, Sapporo Adopts Defense Against Steel Partners ; June 29, 2007, Tokyo Court Deals Blow To Foreign Activist Investors ; August 8, 2007, Japan High Court Keeps Bull-Dog Sauce From Steel Partners' Jaws. Bloomberg News: November 16, 2006, Nissin Food offers $108 million for Myojo stake. International Herald Tribune: July 5, 2007, Steel Partners rejected in Japan, again. Reuters: May 23, 2007, Steel Partners launches offer for Japan Tenryu Saw. FT.com: August 8, 2007, Steel Partners fights on for Bull-Dog ; July 9, 2007, Japan's court approves Bull-Dog poison pill. 29 In Japan it engages in business as Steel Partners Japan Strategic Fund.

49 Chapter 2 M&A activity in Japan

All hostile tender offers in Japan by Steel Partners, the most active company in hostile takeovers, failed. Reuters reports on May 23, 200730 that Steel Partners has spent 337 billion yen to take a stake of more than 5% in 30 Japanese companies.

Livedoor 31 - Livedoor was a fast-growing start-up company that provided internet consulting services to businesses and maintains Web sites for electronic commerce. Nippon Broadcasting System (“NBS”) is a radio broadcaster and affiliated with Fuji Television network, Japan’s top television network. In 2005 Nippon Broadcasting was, owning 22.5% of outstanding shares, the largest shareholder of Fuji Television. The latter decided to launch a tender offer bid for the shares of NBS on 17 January in order to secure management control. In the mean time, Livedoor and its subsidiaries had obtained 35% of the outstanding shares of NBS by 8 February in off-floor trading. The financing of the bid was controversial as Lehman Brothers had given “Livedoor the money in exchange for “death spiral” convertible bonds, which give Lehman the right to convert the bonds into shares at a constant discount to the share price. This, in effect, would reduce the value of Livedoor's shares—indeed, Livedoor's shareholders may already be suffering from that” (The Economist, March 23, 2005). On that day Livedoor president Horie held a press conference in which he declared that he wanted to create a tie-up with the Fuji Television group. The board of directors rejected Livedoor’s proposal the next day and on 23 February NBS decided to issue 4,720 new share rights to Fuji Television. The new issue would amount to 1.44 times the existing shares and reduce ownership of Livedoor’s to 17% and increase ownership of Fuji Television to 59 %. The warrant issue led to a court battle and the ruling did not allow the warrant issue because it was not in the best interest of, or fair to all shareholders. On 7 April Livedoor, NBS and Fuji Television agreed on business collaborations and capital ties. As part of the deal, Livedoor would sell its NBS shares to Fuji Television, in return for which Fuji Television would purchase some 12.75% of Livedoor stocks for 44 billion yen. In September NBS became a wholly-owned subsidiary of Fuji Television.

Yumeshin Holdings32 - Yumeshin Holdings is a construction design company that made a hostile tender offer bid for the majority of the shares of Japan Engineering Consultants (“JEC”), a company that offers consultancy on construction. On 11 July 2005 Yumeshin announced its tender offer period would start on July 20 and end on August 12, aimed at raising its ownership from 6.4% to 53.71%. JEC had taken defensive measures in the form of 5-for-1 stock split if a company would acquire more than 20% of its shares. The use of this split was announced on July 18, and because it was approved by Tokyo District Court, Yumeshin lowered its bidding price to 110.

30 Reuters: May 23, 2007, Steel Partners launches offer for Japan Tenryu Saw. 31 Based on The New York Times: March 25, 2005, Plot Thickens in Japan's Hostile Takeover Battle ; April 14, 2005, Livedoor and Fuji TV in Talks ; April 19, 2005, Hostile Takeover Bid Ends Civilly in Japan. The Asahi Shimbun: February 25, 2005, Livedoor sues to block radio company's tactic. 32 Based on The Japan Times: July 21, 2005, Yumeshin makes bid for Japan Engineering ; July 22, 2005, Yumeshin cranks up takeover bid ; August 9, 2005, White knight rides to consultancy's rescue. Bloomberg: July 30, 2005, Japan Engineering to Issue Options to Thwart Yumeshin (Update1). International Herald Tribune: August 1, 2005, JapanEngineeringaimstothwartbid. Kyodo News International: August 13, 2005, Yumeshin fails in hostile takeover bid for Japan Engineering.

50 Section 2-4 Hostile tender offer attempts

Following this event JEC found a white knight, Eight Consultants Co., a construction consulting company. Eight Consultants made a friendly tender offer at 118 yen per share and explained it would turn Japan Engineering Consultants into a subsidiary. Yumeshin accepted the tender offer and transferred its 22.22% ownership of JEC to Eight Consultants.

Rakuten33 - Rakuten operates the largest online shopping service in Japan and on October 13, 2005 the company announced it wanted to merge with Tokyo Broadcasting System (“TBS”) and create a holding company in order to compete globally. TBS is Japan’s third- biggest television broadcaster and adopted a poison pill defence by offering a unit of brokerage Nikko Cordial an option to buy a stake of up to 21.2% in the event that a hostile bidder would acquire 20% or more of the company’s shares. Rakuten stopped buying shares of TBS after it had acquired 19.09 % of total shares. Mizuho Corporate Bank mediated on November 30 in an attempt to reach an amicable resolution of the situation. TBS agreed to investigate collaboration opportunities and Rakuten agreed to refrain from buying up new TBS shares and to entrust its acquired shares in excess of 10% to a financial institution (Mizuho Trust and Banking Co.).

Don Quijote34 - Don Quijote, a discount retailer, launched a surprise hostile tender offer on Origin Toshu in the period January 16 to February 9. Don Quijote already owned 31.06% and wanted to increase its shareholding to 51% at a bidding price of 2,800 yen. Origin, a food manufacturer and retailer, had not been informed in advance by Don Quijote, strongly disagreed and requested help from Aeon, Japan’s largest department store. Aeon agreed to a friendly counter tender offer for 50.01% at a price of 3,100 yen for the period of January 31 to March 1, 2006. The tender offer of Don Quijote failed but thereafter the company started buying shares of Origin Toshu in the market, announcing on February 15 it owned 47.81% of Origin. At the end of February Don Quijote sold its shareholdings to Aeon that had acquired 95.67 of the Origin by March 1.

Oji Paper35 - The first Japanese hostile tender offer by a blue-chip company was Oji Paper, Japan's largest paper manufacturer, for the shares of Hokuetsu Paper Mills,thesixth largest paper manufacturer. Prior to the bid, Oji Paper had approached Hokuetsu Paper to suggest a friendly merger or acquisition. Hokuetsu Paper objected to the proposal and implemented two anti-takeover devices; it adopted a poison pill and issued 50 million new shares to Mitsubishi Corporation (30.5% stake in the company). Responding to these developments Oji Paper started the hostile tender offer in order to obtain 50%. Hokuetsu Paper did not need to use the poison pill because it had been able to get support from Mitsubishi Corporation, its main bank, and business partners. In addition, Japan’s second

33 Based on International Herald Tribune: October 13, 2005, Online company plans bid for Tokyo broadcaster; October 13, 2005, Rakuten takes aim at Tokyo Broadcasting. South China Morning Post: October 14, 2005, Rakuten moves in on TBS as media shake-out deepens. Nikkei Business: May, 1, 2007, Rakuten asks business heavyweights for help in attempt to form business tie-up with TBS. 34 Based on Reuters: January 30, 2006, Japan's Aeon to make white knight bid for Origin. The Japan Times: February 16, 2006, Don Quijote shocks Origin by boosting stake to 46% ; March 15, 2006, Aeon completes takeover of Origin 'bento' chain. 35 Based on International Herald Tribune: September 5, 2006, Failed takeover bid by Oji seen as loss for Hokuetsu. Wall Street Journal: September 6, 2006, Oji Paper Drops Hokuetsu Bid As Japan Braces for 'Hostile' Era.

51 Chapter 2 M&A activity in Japan

largest paper manufacturer Nippon Paper Group bought 8.8% of Hokuetsu Paper’s shares. Hokuetsu Paper could effectively stabilize 40-50% of the outstanding shares and only 5.3% of the shares were tendered to Oji Paper.

2-5 Conclusion In this chapter we investigated M&A activity in Japan, focussing on mergers and hostile takeovers in the pre-war and post-war periods. For the pre-war period we looked into mergers of companies listed on the Tokyo Stock Exchange in 1937. The number of mergers was low, not exceeding an annual total of 41 merger cases, and merger waves did not occur. We did not find a difference in merger activity of zaibatsu companies and non- zaibatsu companies. Activity was concentrated in the electricity sector, and the main motive for mergers appears to have been the rescue motive as most transactions seem to occur between a strong bidder company and weak target company. Next, we presented M&A activity in the post-war period. Using JFTC data, we showed the number of domestic mergers by annual total and industry and also looked into detailed information on targets for the period 1970-1997. The analysis of the detailed information of each individual year, combined with the Recof information, provided additional insights on the developments in domestic mergers in the post-war period. In this period we found two merger waves: the first in the period 1963-72, and the second from 1991 to present. The first merger wave coincided with the removal of controls on capital transactions in 1964 and appears to be partly influenced by administrative guidance of the Japanese bureaucracy. The largest mergers were concentrated in the Steel, Transportation and Automotive sectors. The main aim of the mergers was to prevent excessive competition and increase the scale of domestic companies in order to become competitive with foreign companies. The second merger wave started after the collapse of the bubble economy and is related to industrial restructuring as the economic recession resulted in bad performance and difficulty to repay loans. In the early 1990s the largest mergers occurred in the Paper and pulp, Chemical and Banking industries, and especially mergers aimed at rescuing failing companies were important. After the regulatory changes in the latter half of the 1990s the number and size of M&A activity increased considerably, although the number of mergers remained stable in the period 1999-2006. In this period we also see that companies were engaging in mergers for strategic purposes. We looked into hostile takeovers during the post-war period and found that the first cases occurred immediately after WWII. During the entire post-war period hostile takeover activity took place, but only in 1999 the first hostile tender offer was initiated. In the period 1999 to July 2007 thirteen hostile tender offer attempts were launched, of which only two were successful. The “hostility” in these successful hostile tender offers can be questioned, however, as the bidder company and target company had a long-standing relationship.

Chapters 3 through 5 will provide background information to interpret M&A activity in the pre-war and post-war period. After a discussion of the informal institutions, formal institutions, and the political and economic development of Japan, we will examine their influence on companies’ financing decision, ownership structure, corporate governance

52 Section 2-5 Conclusion

structure, and formation of corporate groups. This information forms the basis for the institutional model for M&A that we will introduce in chapter 6. According to the institutional model, chapter 7 will subsequently integrate all information and analyze hostile takeovers in the post-war period. In this chapter we explain how the hostile takeovers up to 1999, and the thirteen hostile tender offer attempts in the period 1999-2006 need to be interpreted. In chapter 9 we will investigate the mergers of the second merger wave in Japan empirically. We examine whether shareholder wealth is created, and our analysis focuses on mergers with a rescue motive and the role of the main bank.

53 42

54 3

Informal and formal institutions

This chapter discusses the informal institutions, political development, and formal institutions. These institutions prove to have a major impact on decisions related to financing decisions, ownership structure, corporate governance structure, corporate group formation, and M&A activity. As these decisions are influenced by the historical and context-specific background of managers and companies, we look into the origins of Japan’s informal institutions and describe the political development and formal institutions as of 1639, the year Japan implemented a political system that banned almost all foreign contact. Section 3-1 discusses the informal institutions in Japan, and section 3-2 examines the political development and formal institutions related to financial markets. These institutions strongly influence companies’ financing decisions. Section 3-3 discusses the formal institutions related to ownership structure, corporate governance structure, and M&A activity. We summarize and conclude the chapter with section 3-4.

55 Chapter 3 Informal and formal institutions

3-1 Informal institutions R Characteristics of the (moral) Constitution of Seventeen Articles, which is based on Shinto, Confucianism, and Buddhism, are still present in contemporary Japan. R The corporate ie system is the “house” system that places importance on continuity of a company, rather than its individual members. R The dǀzoku is a group of related “houses” and forms a pyramidal pattern of corporate relationships with personal and financial linkages.

Informal institutions “come from socially transmitted information and are part of the heritage that we call culture” (North 1990, p.37). In this section we look into the development of religion and tradition in Japan. We examine the origins of Shinto and look into Confucianism and Buddhism that entered Japan from Korea and China in 552AD. These three elements had an important influence on the Constitution of Seventeen Articles which was promulgated in 604AD. This constitution of moral development is discussed in sub-section 3-1-2, and the ie system and the dǀzoku system are looked into in the next sub- section.

3-1-1 Religion and tradition Shinto - The indigenous people of Japan is the Jomon that lived in the period from 10,000BC to 300BC. The Jomon was a Mesolithic people that hunted and gathered food in small tribes. Hooker (1996) describes that after they had changed their lifestyle from hunter-gatherers to settlement in large villages, the first signs of religion appeared. In the period 1,000BC to 300BC “the Jomon developed an identifiable religion-they produced a remarkable number of figurines and stone circles constructed outside the main villages begin to appear.”36 Although there was some form of religion present during the Jomon period, the origins of Japan’s current Shinto, and also of the Japanese language and social structure, date back to 300BC (start of the Yayoi period). The Yayoi, a people from the north of China, were the first to enter Japan via Korea and introduced a new religion that would evolve into Shinto. In this period so-called uji developed, clans headed by a patriarch. Each clan had its own god, kami, and these gods were sacred spirits which took the form of things and concepts important to life, such as wind, mountains, trees, rivers and fertility. Religion was transmitted through communal rituals aimed at maintaining harmony between nature, humans and kami. As agriculture had become important for the Yayoi, the rituals became closely tied to the agricultural year. “[A]ll members of the community took part, if only symbolically, in the final meal, thus bringing harmony again to the relationship of humans and the kami”37 (Watt 2003). Battles between the various uji were frequent, and Hooker (1996) explains that “[W]hen one uji conquered another, it absorbed its god into its own religious practices. In this way, the Yayoi slowly developed a complex pantheon of kami that represented in their hierarchy the hierarchy of the uji”. 38 The supreme god of all Shinto gods is the Sun Goddess Amaterasu, and prior to the end of WWII the Imperial Family was regarded as her descendents.

36 http://www.wsu.edu:8000/~dee/ANCJAPAN/ANCJAPAN.HTM 37 http://iis-db.stanford.edu/docs/127/religion.pdf 38 http://wsu.edu/~dee/ANCJAPAN/YAYOI.HTM

56 Section 3-1 Informal institutions

Confucianism - Confucianism entered Japan from Korea and China in 552AD. Confucianism is a tradition founded by Master Kung, K'ung Fu-tzu, from which the Latin name “Confucius” is derived. The most reliable source of his teachings and beliefs is the Lunyu (Lun-yü, or Analects), which is a collection of sayings by and about Confucius and his disciples. Within Confucianism it is believed that to achieve a state of orderliness and peace in the world, it is necessary for human beings to cultivate virtue. In the Analects the concepts of Ren (Jen)andLi are most important. Ren (Jen) can be translated as “benevolence,” “humaneness,” “goodness”, or “virtue”. Li can be translated as “ritual” or “etiquette” and stands for rites, ceremonies, proper behaviour, and good manners. The other three of the five Confucian virtues are righteousness (yi), wisdom (zhi), and faithfulness (xin). Righteousness is doing what is appropriate for one’s role, as for example father, son, teacher, or student. Wisdom has several aspects, including being a good judge of the character of others. Filial piety is acting out of love and respect for one’s parents, and faithfulness includes honesty (Yao 2000, p.34). Yao (2000) indicates that “[C]onfucius repeatedly taught that while it was important to observe ancient rituals strictly, it was even more important to have a sincere heart and a devoted spirit: 'For if a person lacks humaneness (ren) within, then what is the value of performing rituals? For if a person lacks humaneness within, what is the use of performing music?' (Lunyu, 3: 3)” (p.32). In contrast, other Chinese philosophers argued that emphasis should be placed on ritual (Xunzi) or righteousness (Mengzi) (Yao 2000, p.77). Overall, the five virtues need to be studied and trained in order to participate properly in the hierarchies of the two types of human relationships. The first is the ‘three bonds’ (in Japanese: sankǀ), which are functional pairings between rules and minister, father and son, and husband and wife. The second type is the ‘five relations’ (in Japanese: gǀrin)that cover loyalty between rules and minister, filiality between father and son, differential harmony between husband and wife, precedence between elder and younger sibs, and trust between friends (Sekiguchi 2003, p.27). The importance of hierarchical ranking is visible in various elements of the Japanese society. It not only orders relationships of individuals, but also companies, groups, and material objects. There is flexibility in the principle of hierarchy and equality though. When dealing with outsiders, the outside hierarchical relations override the inside hierarchical relations (Hendry 2003).

Buddhism - Similar to Confucianism, Buddhism entered Japan in 552AD. It was founded in India by Siddhartha Gautama, Lord Buddha, who lived from 560BC to 480BC. His teachings were not written down but transferred from one generation to the next by word of mouth, which led to various groups with their own interpretation. The main teaching of Buddhism is the path of practice that an individual can take up to gain release from suffering. Most important in this respect is that an individual needs to try to overcome self- centeredness and have compassion for fellow beings. “[T]he Buddha held that to gain release from suffering one had to attain a new understanding of reality. In particular, one had to see that persons and things do not exist autonomously on the basis of individual “selves,” but rather that all things are linked in a network of interdependency. To overcome the self-centeredness of the ignorant, one had to transform one’s way of thinking and acting through the practice of Buddhist morality, mediation and wisdom or study” (Watt 2003).

57 Chapter 3 Informal and formal institutions

3-1-2 Constitution of Seventeen Articles By 500AD the uji had evolved into military states headed by aristocratic families who were constantly fighting over power. The situation of constant struggles between uji led to the promulgation of the Constitution of Seventeen Articles in 604AD by Shǀtoku Taishi, a regent under Empress Suiko (ruling between 592AD-628AD). The document was not a legal constitution, but stressed virtues of harmony and importance of moral development for government officials, based on the principles of Buddhism and Confucianism. The following year he also implemented court hierarchy and “twelve grades of rank were established, named after six virtues, subdivided into greater and smaller” (Sansom 1978, p.91). The Constitution of Seventeen Articles formed the basis for the Taika Reform in 645AD, which was primarily aimed at unifying the country to be able to levy tax and defend itself from foreign intrusions. The reform converted uji from separate states into provinces and gave the Tennǀ, “Divine Emperor”, absolute power over the provinces. Herewith Japan was changed from “an uneasy confederation of tribes into a centralised bureaucratic state on the Chinese model” (Sansom 1978, p.95). The new position of the emperor was in conformity with the principles of both Shinto and Confucianism. According to Shinto, the Emperor was a legitimate descent from the Shinto sun goddess Amaterasu and “ruled by theDecreeofHeaven”39 (Hooker 1996). Confucianism legitimates the position of the emperor as the parent of the people and patriarch of the family state, based on the principles of patriarchal family and filiality. Sekiguchi (2003) explains that if “filiality in the family is extended to the public realm, it becomes loyalty, the basis of public morality. Loyalty to the family patriarch is thus the foundation of state rule” (p.29). The heritage of the Constitution of Seventeen Articles was not only the basis for the Taika Reform, but its characteristics are still present in contemporary Japan. Below we describe the articles we think are important in relation to our discussion on M&A in Japan. We focus in particular on the rules of conduct that are likely to influence decisions related to ownership, corporate governance, and M&A activity; subjects that will be examined in the next chapters.

Harmony & Tolerance - The Japanese accentuate harmony and human nexus, and they try to avoid conflict, confrontation, and criticism. Nakamura (1958) mentions about the attitude towards the co-existence and unity of Shintoism, Buddhism, and Confucianism: “[A]mong the scholars conflicts and controversies were carried on which advocated different doctrines, but among the common people they were made to fuse and harmonize. The Japanese are by nature inclined to rapprochement without threshing out an issue” (pp.299-300). A good example is the previously discussed way that kami of conquered uji were absorbed in the conqueror’s religious practices. Other examples of tolerance are the absorption of foreign words and foreign cultures. The Japanese welcome these cultural elements from other countries by integrating them in their own culture. In this process, however, they will never deviate from what has been inherited from the past such as, for example, their ancestral gods.

I. Harmony is to be valued, and an avoidance of wanton opposition to be honoured. All men are influenced by class-feelings, and there are few who are intelligent. Hence there

39 http://wsu.edu/~dee/ANCJAPAN/YAMATO.HTM

58 Section 3-1 Informal institutions

are some who disobey their lords and fathers, or who maintain feuds with the neighboring villages. But when those above are harmonious and those below are friendly, and there is concord in the discussion of business, right views of things spontaneously gain acceptance. Then what is there which cannot be accomplished ! X. Let us cease from wrath, and refrain from angry looks. Nor let us be resentful when others differ from us. For all men have hearts, and each heart has its own leanings. We are not unquestionably sages, nor are they unquestionably fools. Both of us are simply ordinary men. How can any one lay down a rule by which to distinguish right from wrong ? For we are all, one with another, wise and foolish, like a ring which has no end. Therefore, although others give away anger, let us on the contrary dread our own faults, and though we alone may be in the right, let us follow the multitude and act like them.

Modesty & Group - In Japanese interpersonal relationships, modesty about oneself is highly valued, and compliments should be given to others.

VI. Chastise that which is evil and encourage that which is good. This was the excellent rule of antiquity. Conceal not, therefore, the good qualities of others, and fail not to correct that which is wrong when you see it.

It is important to be reserved about one’s own desires because the group is more important than the individual. Individual attitudes are subordinated to the group’s interest, which implies that people individually do not express their own ideas. If there is one specific idea within the group that everyone agrees to, individuals will not move against the group’s idea.

XV. Turn away from what is private, and to set our faces towards that which is public- this is the path of a Minister. Now if a man is influenced by private motives, he will assuredly feel resentments, and if he is influenced by resentful feelings, he will assuredly sacrifice the public interests to his private feelings. When resentments arise, it interferes with order, and is subversive of law. Therefore in the first clause it was said, that superiors and inferiors should agree together. The purport is the same as this.

The importance of the group can be seen in decision-making as well. An example of the importance of the group is the ringisei. According to this system a document is circulated to managers whose department is affected by a particular decision. Each manager signs the document indicating assent to the decision, the document functioning primarily as developing consensus within the company.

XVII. Decisions on important matters should not be made by one person alone. They should be discussed by many. But small matters are of less consequence. It is unnecessary to consult a number of people. It is only in the case of the discussion of weighty affairs, when there is a suspicion that they may miscarry, that one should arrange matters in concert with others, so as to arrive at the right conclusion.

Trust - Trust or good faith is another very important norm in relationships between Japanese.

59 Chapter 3 Informal and formal institutions

IX. Good faith is the foundation of right. In everything let there be good faith, for in it there surely consists the good and the bad, success and failure. If the lord and the vassal observe good faith one with another, what is there which cannot be accomplished ? If the lord and the vassal do not observe good faith towards one another, everything without exception ends in failure.

3-1-3 Ie system The ie system is often translated as ‘family system’, but as Hendry (2003) suggests, the actual meaning of the system is better reflected in its literal translation ‘House’ because of its “connotation of continuity” (p.26). The importance of the ie system lies in the ie being the continuing entity, rather than any individual member. Conversely, the continuity of the ie is most important for each individual member who is expected to have complete loyalty towards the ie. The relations within the ie are characterized by Confucian principles, the head is responsible for all his subordinates and enjoys privileges given by the subordinates. Hendry (2003) indicates that this is not an independent position and that “if any particular head became despotic or detrimental in some other way to the house as a whole, he could be removed according to the decision of a wider family council” (p.27). The hierarchical lines within the ie are organized based on age, sex, and expectation of permanency in the house. Although it was customary and desirable that the eldest son would succeed his father, this was not essential. The continuation of the ie was most important and in a family business this was stable management. Regarding the importance of stable management, Ito (1997) indicates that this does not necessarily imply business expansion but “[R]ather, as is often stated in the family precepts of merchant families, the emphasis is on thoroughgoing service and commodity quality, for which the maintenance of appropriate management is a prerequisite; and it was for this reason that importance was attached to such material requirements as land, family residence, and store premises” (p.35). Regarding succession, if the eldest son was not qualified as a successor of the head or “[I]f there were no children at all, a new spouse could be sought, or a relative’s child could be adopted, or the head could take a concubine to produce an heir, who would then be brought up by his wife. It was also permissible, if necessary, to adopt a totally unrelated child, so that the blood connection, while desirable, was not indispensable to the continuity of the ie” (Hendry 2003, p.28). If there were only daughter(s), the spouse could be selected based on his qualities and this was often most preferred by the head of the ie (Kitano 1962). A son that did not become the heir could set up his own house and start an ie, or marry into another ie. The latter option meant he had to leave his own ie. If he set up his own house, this would be regarded as a branch of the main house and this could develop into a wide group of houses. This group of related houses is called a dǀzoku and is formed as a pyramidal pattern of relationships with personal and financial linkages. Within the dǀzoku the junior houses are ranked hierarchically in order of genealogical distance or in chronological order of being hived off from the parent house (Cornell 1964, p.454). Both the relations within the dǀzoku as well as those with companies outside the dǀzoku are based on Confucian principles. The companies in the hierarchical pyramid have a particular position or rank in relation to other (group) companies, and employees of the company share the same status as the company in the outside world. The ranking of the company in relation to other companies is very

60 Section 3-1 Informal institutions

important and the employees are driven to improve the company’s status as to advance their own status (Hendry 2003). Interaction between companies can develop and evolve into long-term and stable relations and even oyabun/kobun relations. Oyabun means parent-part and kobun means child-part; with reference to companies it refers to the relation of a parent company with a subsidiary company (Hendry 2003, p.119). The oyabun/kobun can also develop regarding the relation of an employer and employee. Caldarola (1965) argues that when industrialization of Japan started, the factory managers had difficulty to recruit people because of the family-orientated psychology: In fact, for people nurtured on the family system there is great inner need for this basis that is a guarantee of subsistence, a source of relief, a support to one’s spirit against disorder and insecurity of the world outside. A certain guarantee of the extension of paternalistic ties made it possible for youngsters to venture away from the shelter of home (pp.367-368). This led factory managers to create a safe haven in the form of a new family, the “company family”. The company takes responsibility for the well-being of the employee and the employee will be loyal to the company. This type of oyabun/kobun relationship is not only between employer and employee, but also teacher-student, parent-child, and between the senpai and kǀhai within Japanese companies. The senpai is a person who is already working in the company when the kǀhai enters. The senpai feels responsible for the development of the kǀhai within and outside the company. He teaches the kǀhai how to function in the company and gives assistance and advice. The kǀhai on his part builds up debt to the senpai and repays this with loyalty and respect towards the senpai (Cole 1971; Hendry 2003). Lifetime employment, seniority-based wages, and enterprise union are three important elements of the labour system in Japan, originating from the “company family” system. Lifetime employment is the system in which employees are recruited directly from school or university and remain working for the same company until retirement. The seniority wages system entails that salary progression is based on length of service with the company, rather than the individual’s skill or output. Milhaupt (2001) describes this as “a mutual expectation of employer and employee […] that the employee will enjoy continuous employment until mandatory retirement age. Wages that undercompensate for productivity at the outset of the employment relationship are eventually offset by wages that overcompensate for productivity in the latter stages of the employee’s career” (p.8). The enterprise union is a union for all employees in the company up to middle- management, which enhances identification with the company's interests by all employees (Crump 2003, p.5). Another important aspect of the “company family” is that it not only concerns the company, but also the group to which the company belongs. Starting in the pre-war period, employees were assigned to jobs in related group companies. One of the reasons was to create a vast understanding of the group’s business activities, but more importantly, to build and strengthen the relationships between employees and companies. Kester (1989) argues about the post-war job rotation that employees will treat each other with respect because in the future they might work together.

61 Chapter 3 Informal and formal institutions

3-2 Formal institutions R The pre-war Japanese government tried to set up a modern financial and legal system in order to be able to renegotiate the unequal treaties it had signed in 1857-58. R A modern banking system, stock market, and bond market were set up and functioned effectively up to 1937. R After WWII, the Supreme Commander for the Allied Powers’ policies were aimed at (i) demilitarizing and democratizing Japan, and (ii) setting up a regulatory environment for a safe and stable financial system.

In this section we look into political change, regulatory changes, and the development of financial markets from the start of the Meiji period (1868-1912) to present. In order to understand the (importance of the) changes in the Meiji period, we first describe some aspects of the preceding period, the Tokugawa period (1600-1868). Next, we investigate laws and regulations related to the banking sector, bond market, and stock market.

3-2-1 Pre-war political development and financial markets As of 1639 Japan maintained a system of sakoku, 'the closed country', and banned almost all foreign contact for more than two centuries. Only diplomatic relations with Korea and the Ryukyu Islands (Okinawa) were upheld, and trading relations were strictly controlled and restricted to those conducted with Chinese and Dutch merchants. Japanese were forbidden to leave Japan, and foreigners were not allowed to settle in Japan (Allen 1946). The political system was formed by the bakufu (central government) and the han (local government). The bakufu, headed by the shǀgun, had absolute political power over the han that were ruled by daimyǀ. The Japanese society was sociologically separated into four classes: the samurai, peasants, artisans and merchants (shinǀkǀshǀ). In this system the samurai had the highest status as soldier and ruler. Peasants were awarded the second highest class as they were the foundation of the state through their production of rice and payment of taxes. Artisans were placed at the third class and merchants formed the lowest class. Artisans were ranked above merchants because they provided the samurai class with goods such as weapons and military supplies; merchants were primarily engaged in pursuit of profit and self-enrichment. Individuals were supposed to remain in, and marry within, their own social class (Hendry 2003; Beasley 1963; Sheldon 1958). In the nineteenth century the system of sakoku started to become diffult to maintain as industrialized Western countries expanded trading activities in East Asia. Commodore Perry entered Tokyo Bay with four military ships in 1853 and demanded Japan to open its ports. In 1857 and 1858 the bakufu was forced to open the country to foreign trade and signed unequal treaties with Britain, the U.S., France, Holland, and Russia. The unequal treaties resulted in a severe loss of Japan’s sovereignty as foreigners were free to settle in designated areas and granted legal extraterritoriality. In the treaties Japan had also renounced its tariff autonomy, resulting in marginal or no tariffs on foreign goods from the countries involved. The treaties raised violent political opposition among the lower rank warriors of some han (Satsuma, Choshu, Tosa and Hizen), who were dissatisfied. As the shǀgun had signed the treaties without consent of the emperor or other daimyǀ,the warriors of the four han engaged in a coup d'état in the name of patriotism. They seized power, established the Meiji government in 1868, and restored the Emperor to a constitutional position (which he had not occupied for 800 years). The contact with foreign

62 Section 3-2 Formal institutions

countries and the unequal treaties resulted in a transformation in Japan’s economy, legal system, and economic regulations. The fear of becoming colonized by a Western country, led to the Meiji policy of fukoku kyǀhei (wealthy nation, strong army) (Fukasaku 1992; Beasley 1963, p.2; Allen 1946, p.25). The feudalistic system of the Tokugawa period was changed into a centralized bureaucracy in 1869, and clans surrendered their domains to the government. In 1871 local government in han was replaced by prefectures (haihan chiken),andduringtheperiod 1870-71 the four-class system was abolished (shimin byǀdǀ). In this period the old financial and administrative system associated with feudalism disappeared, and homogenization of citizens (kokumin) and establishment of a centralized nation-state (kokka) was promoted (e.g. Suzuki 1996, p.27). In 1889 the Meiji constitution was promulgated and the following year the Diet was opened, replacing the political system with a centralized government and prefectures. Development of financial markets started as the government committed itself to building a modern financial system organized around a modern banking sector. The financial system and financial markets followed the market mechanism and developed rapidly until 1937. After 1937 the financial system changed as the government took control over funding of companies. According to the Temporary Funds Adjustment Act the government allocated funds to war-related industries. The 1938 National General Mobilization Act gave the government even more control over the economy as it authorized decisions on aspects such as use of labour and materials (e.g. Röhl 2005).

Banking sector - The National Bank Act of 1872 was the first step towards the creation of a modern banking system. Modeled after the National Bank Act in the United States40,the system was decentralized, and national banks regulated the money supply. Each national bank had to hold 40% of equity in gold and could issue bank notes at 60% of its equity. Because bank notes had to be convertible into gold, only four banks had been established by 1874. The National Bank Act was amended in 1876, and the reserve requirement for gold was relaxed to 20%; the remaining 80% was allowed to be national bonds or samurai stipends. Notes could be issued up to 80% of equity and needed to be convertible into government notes. By 1880 the number of national banks had increased to 151 (Teranishi 2005). In the years 1876-1879 the amount of bank notes increase from 1.7 million yen to 34.4 million yen, causing an annual inflation rate of 20% (Hoshi and Kashap 2001). The high number of national banks and the absence of a monetary policy led to high inflation and eventually to the abandonment of this system. A central bank, the Bank of Japan, was established in 1882 and, according to the Convertible Bank Notes Act (effective as of 1884), became the only entity to issue the national currency. Similar to Germany, a system was set up that separated the type of funding by financial institution. The national banks were gradually transformed into commercial banks, and the Hypothec Bank of Japan and Industrial Bank of Japan were established to provide long-term credit to the sectors agriculture and industry (Allen 1946, p.47). In 1899 all national banks had become ordinary banks, and the total of ordinary and saving banks increased to 2,334 in 1901 (Teranishi 2005, p.59, table 3.1). In 1901 the capital requirement for joint stock banks was set at 500 thousand yen and for private banks

40 This US National Bank Act was enacted in 1864, did not have a central bank but national banks that were authorized to issue bank notes.

63 Chapter 3 Informal and formal institutions

at 250 thousand yen, encouraging consolidation of banks. By the end of 1913 the number of banks had fallen to 2,158. Due to the banking crises of 1920, 1922 and 1927, smaller and weaker banks fell into bankruptcy or were acquired by larger banks.41 The Ministry of Finance also restricted the establishment of new branches in 1923, and this prompted large banks to acquire smaller banks in order to expand their networks. The Trust Act and Trust Business Act, effective as of 1923, increased the activities of banks as they were allowed to engage in the placement and underwriting of bonds. The revised banking law of 1928 raised the capital requirement of banks in Tokyo and Osaka to two million yen and of other banks to one million yen. This further stimulated mergers between smaller banks and absorption of smaller banks by larger banks. The number of commercial banks fell from 1,515 to 65 in the period 1928-45, and a high proportion of the total banking business came to be conducted by a very few large banks working with branches. By the end of 1928 the “big five” banks (Mitsui, Mitsubishi, Sumitomo, Yasuda and Dai-ichi) held 34% of the deposits of the ordinary banks, and if the deposits of the nine next largest ordinary banks are added, this proportion increases to 55%. The first four of the “big five” banks were part of zaibatsu groups, and these groups greatly increased their financial power during the 1920s (Allen 1946; Hoshi 1995; Okazaki 2006).

Stock markets42 - Whereas the banking system was imported from the West, the stock markets originated from the rice and commodities markets that had developed during the Tokugawa period. The government failed to introduce the rules of the London exchange in 1874 because the private sector felt these were too restrictive. In 1878 it “promulgated a new Stock Exchange Act, which recognized many transaction formats inherited from rice markets” (Hoshi and Kashap 2001, p.25). The first stock exchanges, the Tokyo Stock Exchange and Osaka Stock Exchange, were also established in this year. Okazaki et al. (2005) show that the number of exchanges peaked in 1897 with 46 exchanges, but declined thereafter and remained at around 10 exchanges from 1910. The Tokyo Stock Exchange was the largest stock exchange in terms of revenue in the pre-war period.

Bond market - Ohno (2006) argues about securities exchanges that “initially, few stocks were traded and these exchanges functioned mainly as a secondary market for government bonds. Former samurais who received government bonds in exchange for the previous rice salary often wished to sell them as they faced financial distress. In the 1880s, as many railroad companies were established, railroad bonds gradually became the most important instruments for trading” (p.92). According to the Commercial Code of 1890 it was required that the name of the owner was inscribed on bonds, and this did not make them attractive investments. The Commercial Code was amended in 1899, and allowed companies to issue bearer bonds (Hoshi and Kashap 2001). In 1905 the Secured Debenture Trust Act was passed and “allowed large banks to be trustee of the collateral securing corporate bonds on behalf of the owners of securities” (Hoshi and Kashap 2001, p26). This, however, did not result in an increase of bond issues abroad. Bond financing increased after 1920 and especially in the years 1928, 1934, and the period 1939-45 it was an important way of financing.

41 Please refer to section 4-1-1 for more information on the economic development during this period. 42 See: Suzuki (1989), Kaizuka (1986), Oka (1993) and Akimoto (1993)

64 Section 3-2 Formal institutions

3-2-2 Post-war political development and financial markets After WWII, Japan was placed under the command and authority of the Supreme Commander for the Allied Powers (“SCAP”). SCAP’s main concern was to demilitarize and democratize Japan, and to set up a regulatory environment in order to create a safe and stable financial system. These regulations intended to supply sufficient capital to developing industries and to enhance the availability of funds in low productivity and traditional areas. Japan’s new constitution was enacted on 3 May 1947 and changed Japan from a centralized bureaucratic state into a democratic country with a cabinet system. The occupation of Japan and supreme power of SCAP came to an end when the Peace Treaty took effect on 28 April 1952 (Wagatsuma 1956, p.107). The Imperial Order 567, set forth in November 1946, prohibited companies that had been part of the same zaibatsu group prior to the war from holding each other’s stocks. Until the Peace Treaty this “completely prevented the same-line companies from holding the liquidated (by zaibatsu dissolution) and newly issued stocks” (Miyajima 1994, p.320). The Antimonopoly Act was passed in 1947 and prohibited (i) private monopolies to control business activities and restrain competition, (ii) unreasonable restraint of trade, including cartels (with some exemptions, as will be explained in sub-section 4-1-2), and (iii) unfair trade practices. Another important aspect was the prohibition of the pure holding company in the post-war period. The implemented measures included forced liquidation of 42 holding companies and the two largest trading companies, Mitsui Bussan and Mitsubishi Shoji, confiscation of stock-ownership of zaibatsu families, purging of 1,575 top executives of more than four hundred companies, and the prohibition of the use of the pre- war trademarks by Mitsui, Mitsubishi, Sumitomo, and Yasuda (Odagiri 1992, pp.171-172). The U.S. concern for the Soviets and their influence in Korea changed their attitude to rebuilding Japan as an economic ally of the United States; the initial program of anti-trust legislation and de-concentration measures was largely cancelled. The change in the U.S. attitude is especially visible with reference to the de-concentration and reorganization activities of Japanese companies. Initially 257 industrial companies and 68 companies in distribution and services had been designated for reorganization, but the final plan of 1949 involved only 11 companies for dissolution (excluding Mitsubishi and Mitsui trading companies that had already been dissolved), and 7 companies for divestiture (of plants or shares). After the peace treaty came into force in 1952, the zaibatsu reassembled the following decade as keiretsu but, as we will explain in section 4-3, their form and substance was substantially different (Johnson 1982). The Securities and Exchange Law (shǀken torihiki hǀ) was enacted in 1948 and, inspired on the Glass-Steagall Banking Act in the United States, it regulates the relationship between the stock company and capital markets. It also re-established the structure of segmented financial markets and specialized institutions. Securities business was only to be performed by securities companies, and commercial banks were responsible for short-term financing. Long-term financing was left to long-term financial institutions, such as trust banks and long-term credit banks. Compared to the pre-war financial system, commercial banks and trust banks lost the right to engage in underwriting for bonds. Within the banking sector the so-called convoy system was implemented. This system refers to a wide set of financial regulations that limit competition, including interest rate controls, branching restrictions, and restrictions on the range of financial products available (Teranishi 2005; Hoshi and Kashap 2001).

65 Chapter 3 Informal and formal institutions

Deregulation of the Japanese financial system became an ongoing process as of the 1970s. In the latter half of the 1970s bond markets were gradually deregulated, and the Foreign Exchange and Trade Control Act was reformed in 1980. After the reform companies were allowed to turn to foreign markets for funding and making investments abroad. Other changes in laws and regulations during the 1980s were characterized by efforts to establish a financial environment that was more responsive to competition. In this decade the financial reform increased competition not only among banks, but also between banks, non-depository institutions, and other private financial institutions such as securities companies. This new situation made the financial institutions also look for other business opportunities. According to Steenbeek (1996), even until the end of the 1980s the primary focus of the Japanese regulators was insuring the soundness and stability of the financial system through government protection of the domestic financial institutions. Only during the 1990s the financial institutions were forced to give up some protection to compete with institutions out of reach of Japan's regulators. This policy change, enforced by the changing international financial markets, was pursued during the 1990s and can be seen in the Big Bang-plan.43 The Financial System Reform Act that went into effect in 1993 meant a change in the separation principle within the financial system. Banks were allowed to form subsidiaries that could operate in the securities business, and securities companies were allowed to set up subsidiaries to engage in trust business. In November 1996 the new financial deregulation plan “Japanese Big Bang” was announced. The main aim of the plan was to make the existing Japanese financial system free, fair, and global. Regulators hoped to use the Big Bang to liberate Japan from the long-lasting economic slump as it promoted innovation in financial products and services, and improved the openness of the entire system. The Japanese government realized that its measures of protection of the banking sector needed to be abandoned. The convoy system caused the large moral hazard problems with banks and had resulted in non-performing loans at financial institutions. In 1997 the severity of the problems of Japanese financial institutions became evident in bankruptcies of the Hokkaido Takushoku Bank and Yamaichi Securities Company and the Long-Term Credit Bank of Japan and Nippon Credit Bank in 1998 and 1999, respectively. Large Japanese banks were faced with rising Japan premiums and lower credit ratings due to high amounts of (non-performing) loans. This situation, combined with the difficulty to maintain the level of capital requirements of the Basel Committee to operate internationally, resulted in banks selling shareholdings in companies (Miyajima and Fumiaki 2005, p.14; Scher 2001, p.1).44

Stock market - The Securities and Exchange Law was enacted in March 1947, and in April 1949 stock exchanges were re-established in Tokyo, Osaka and Nagoya. Trading on these exchanges began on 16 May. Until the late 1960s the standard for equity issue of stock was par instead of market value. As the market value was typically higher, this discouraged companies from turning to equity financing. Combined with the ("required") dividend payment at 10% of face value, it was a relatively expensive form of funding from a cash flow perspective as well. Stock issues at market price were permitted from 1966

43 See: Ito (1997) and Nihon Keizai Shinbun (1997) 44 We will show the implications on companies’ financing decision and ownership structure in sub-section 4-3-4.

66 Section 3-2 Formal institutions

onward, but only during the 1970s public share issues started to become popular. The difference with the years just after 1966, when the par value-restriction was abolished, was the rise in market prices during the 1970s (Hoshi and Kashap 2001). During the 1980s and 1990s the development of the stock market is characterized by further financial deregulation. The most important effect of the deregulation was that companies could more easily enter foreign stock markets as the strict rules were gradually abolished.

Bond market - Because of the balanced budget doctrine in the high growth period, the Japanese government did not issue bonds for revenue purposes from the mid-1950s until 1965. In 1965 the government began to issue seven-year fixed coupon bonds for the first time. Since the mid-1960s government bond issues have dominated the bond market, with corporate bonds representing only a very small fraction of the total market. Even in the period 1986-90, straight corporate bond issues accounted for a mere 3% of total bond issues in Japan (Hodder and Tschoegl 1993). The reason why Japanese companies have not used bonds to obtain financial resources lies in the collateral requirements and issue terms. Corporate issuers had to pay management fees and underwriting commissions, and compensate a trustee for a variety of services that substantially increased issue costs. Also, companies had to obtain approval on the terms and timing of the issues from a committee dominated by a group of large banks (referred to as "commissioned banks"). This procedure was cumbersome and lacked flexibility. Only as of 1979, Japanese companies were allowed to issue unsecured bonds. And initially, only a limited group of companies was eligible for making these issues. The restrictions on bonds were relaxed in stages, until several hundreds of companies were eligible for unsecured issues as of 1988, and the number of bond transactions increased. So, until the late 1980s the bond markets were either closed or prohibitively expensive for most Japanese companies. In the 1990s, the importance of bond issues has been increasing, but the issues primarily involved foreign bonds.

67 Chapter 3 Informal and formal institutions

3-3 Formal institutions related to M&A R Before WWII the general shareholders’ meeting had unlimited authority in all financial and managerial matters and appointed the board of directors and auditors. R After WWII the minority shareholders gained a lot of power, and authority on managerial matters was transferred from the general shareholders’ meeting to the board of directors. R As of 1997 companies can set up pure holding companies and merger procedures have been simplified. R As of 2001 companies can freely buy back their own shares and cancel or hold them as treasury stock.

Table 3-1 shows the major legislative changes, announced and implemented since the late 1990s, which created various incentives for Japanese companies to engage in M&A activity. Important changes include amendments to the Japanese Commercial Code (JCC), the enactment of the Industrial Revitalization Law, and implementation of new accounting standards. This section describes the formal institutions that have a direct influence on M&A activity. Our analysis of the company law goes back to its origins in the pre-war period, and we discuss (the amendments of) the Antimonopoly Act and securities exchange law for the post-war period.

Table 3-1 Important laws and regulations related to M&A as of 1997 Holding companies 1997 Simplified merger process Stock option system 1999 Stockswapandstocktransfer New accounting system Industrial revitalization law 2000 Civil Reoganisation Law 2001 Easier procedure for spin-offs and split-offs Freely repurchase own shares 2005 Amendments simplified merger process Short-form merger process 2007 Merger currency flexibility

3-3-1 Company Law A very important role in the modernization of the Japanese economy in the Meiji period was the introduction of new company structures. The joint-stock company was introduced in the 1870s and strongly promoted by the government. As previously described, companies were traditionally organized through the ie system and capital was raised from within that system. The introduction of the joint-stock company created the possibility to raise capital from anonymous investors through share issue. Due to the lack of a legal environment, however, the introduction of the joint-stock company structure created various problems. The most serious problem was that due to “the lack of regulations there was no sufficient limitation of liability of shareholders in case of insolvency. Against all their expectations, the shareholders were held responsible for the losses, with dire consequences for their private assets” (Röhl 2005, p.352).

68 Section 3-3 Formal institutions related to M&A

In July 1893 the Company Law, Bills and Notes Law, and the Bankruptcy Law of the Old Commercial Code (kyǌ shǀhǀ) came into force and solved some of the problems that companies and investors faced. The Old Commercial Code introduced limited liability for shareholders, a license system that necessitated joint-stock companies to get governmental approval for incorporation, and introduction of required bodies within companies. Regulations regarding the bodies prescribed the joint stock company to consist of a general shareholders’ meeting (sǀkai), a board of directors (torishimari-yaku), and auditors (kansa- yaku). The general shareholders’ meeting had unlimited authority in all financial and managerial matters and appointed the board of directors and auditors. A minimum of three directors was mandatory and each individual director could legally represent the company. Auditors supervised directors’ behavior, assessing whether it was lawful and in line with the shareholders’ interests, but could not dismiss directors. It was possible for auditors to be board members at the same time Blakemore and Yazawa 1953, pp.16-17; Tatsuta 2005, pp.192-193). In 1899 the Old Commercial Code was replaced by the New Commercial Code. The New Commercial Code changed the licensing system for stock companies into a registration system and provided regulations for mergers. The Old Company Law did not have any provisions for mergers and if two companies wanted to merge, one of them had to be dissolved first. Another amendment in the new law was that directors had to be shareholders, and auditors were not allowed to act as directors simultaneously. Effectively, this meant that a lot of companies were managed by large shareholders as there were comparatively few small investors in most companies. A large shareholder also often took the role of the auditor (Röhl 2005, pp.350-362). The amendments in 1911 were primarily aimed at preventing the creation of so-called “bubble companies” (hǀmatsu kaisha) by introducing strict personal liability for promoters. The limited liability of stock companies invited a lot of well-known persons, promoters, to establish companies based on their reputation, selling stock to investors at a substantial premium. “[O]ften these promoters were mostly interested in cashing up that premium and did not actively engage in running the company. Usually, the result was a prompt business failure with the investors losing all their invested capital. A further consequence was an increasing loss of confidence of the investing public in the stock corporation in general” (Röhl 2005, p.374). The amendments in 1938 improved the protection of shareholders by providing better means to control the board of directors, increasing the number of cases for which shareholder approval was required. The requirement of being a shareholder to become a director or auditor was also abolished and made it easier for companies to recruit managers based on the required skills. The board of directors could select directors who were allowed to represent the company by themselves (Röhl 2005, pp.373-376).

Post-war period - After the war the company law was amended by SCAP based on the U.S. Uniform Stock Transfer Act of 1909 and Illinois Business Corporation Act of 1933. The changes in the post-war company law related to the reorganization of corporate powers, the strengthening of shareholder rights, and the creation of new financing opportunities (Blakemore and Yazawa 1953). The zaibatsu dissolution and subsequent dispersed shareholding in companies entailed that the shareholders’ general meeting did no longer function as an efficient manager. This necessitated a reorganization of corporate powers and authority was transferred from the shareholders’ general meeting to the board of directors. The authority of the shareholders’ general meeting was limited to matters such

69 Chapter 3 Informal and formal institutions

as amendment of the articles of incorporation, decrease in stated capital, decision on the company’s consolidation, approval of financial statements, declaring dividends, and election of directors and auditors. The board of directors became the body responsible for managing the company and strengthened its position. For example, the board of directors had the power to issue additional shares up to the ceiling authorized by the articles of incorporation. In contrast to the individual powers of pre-war directors, directors could only control the company collectively. One or more representative directors was/were appointed to take charge of daily operations. The power of corporate auditors was limited to auditing of financial statements and reporting to the shareholders’ meeting. As part of the pursuit of corporate democracy by SCAP and to protect minority shareholders, the power of individual shareholders increased as a compensation of the loss in power of the shareholders’ meeting and auditors. Various restrictions on voting rights were abolished and shareholders were given direct controls over the board of directors in case of unfair or illegal business practices. Any shareholder who had been a shareholder for six months could initiate an action to enforce resignation of a director. “[T]o ensure that these rights and obligations could be enforced, a derivative suit was introduced following again the American model. To enhance the property rights of the shareholders, restrictions on transferability were abolished” (Röhl 2005, p.394). The company law was revised many times in the post-war period. As a consequence of business failures due to window-dressing of financial statements in the mid-1960s, a major reform was implemented in 1974. It was mandated that large companies had their financial statements audited by certified public accountants prior to approval by the general shareholders’ meeting. As of 1981 large companies were required to appoint at least two auditors (Tatsuta 2005, p.195).

Table 3-2 Ownership and rights in post-war Japan Ownership Rights (voting rights) 1.00% • Right to propose agenda at shareholders' meeting 3.00% • Right to request the calling of a shareholders‘ meeting • Right request court to review majority decision not to remove a director • Right to demand court-supervised inspection of company’s operation and finances 10.00% • Right to request court to review proposed dissolution of the company > 33.3% • Blocking minority. Negative control – ability to veto special resolutions • Majority control - ability to pass ordinary resolutions including in relation to >50.0% appointment of directors and auditors > 66.6% • Full control - ability to pass special resolutions in relation to various issues including - Removal of board members - Transfer of the company's business - Stock-for-stock (stock swap) -Stocktransfer -Merger

70 Section 3-3 Formal institutions related to M&A

Since WWII it is possible for shareholders with a one percent ownership stake in a company to propose the agenda on the shareholders’ meeting. 45 With three percent ownership a shareholder can request a shareholders’ meeting, ask the court to investigate why the majority decided not to remove a director, and inspect and make extracts of the corporate books and records. 46 An ownership percentage of ten or more gives the shareholder the right to request a court review of the dissolution of a company. If a shareholder owns one third of a company’s voting rights, he can block special resolutions such as M&A transaction and amendments to the articles of incorporation, and so on. Ownership of more than 50% gives the owner the right to pass ordinary resolutions such as electing and dismissal of the board of directors. 47 A shareholder can pass special resolutions with ownership of more than two-third of the outstanding shares.

Stock buyback - Japanese companies had been prohibited to acquire and own their own shares until 1994 when the Commercial Code was amended. As of 1994 companies are allowed to acquire their own stock (treasury stock) “for legitimate reasons (such as for transfer to an employee stock ownership program), when transferred to employees” (Jetro 2001, p.4). Stock option programs were authorized after the revision in 1997 and provided flexibility in employee compensation schemes and incentives for managers to consider share price. The option of the ‘own stock’ type enables a company to buy back shares and then transfer these to directors and/or employees in the form of stock options. The ‘stock warrant type’ gives directors and/or employees the right to receive new stock (Shishido 2004, pp.12-13; Jetro 2001, p.4-5). As of 2001 companies can freely acquire their own shares and cancel or hold them as treasury stock. Shishido (2004) indicates that “share repurchases become basically unrestricted within the bounds of profit payable as dividend and subject to a few procedural rules. Even the resale of treasury stock was deregulated” (p.10). Revision of the law in 2003 allows the board of directors to decide buyback timing and size once required amendments of the articles of incorporation have been approved at a general shareholders’ meeting.

M&A - Deregulation of merger procedures was started in 1997 and procedures were simplified as follows: (i) the requirement for approval of the merger by one of the two shareholder meetings was abolished, (ii) it was no longer necessary to notify individual creditors, and (iii) simplified mergers were permitted. In a simplified merger the bidder

45 Prior to the 1981 reform this was possible for shareholders that, individually or as a group, held 1% of the outstanding shares or 300 shares during six months or longer. In 1981 listed companies bundled their shares into units, with the unit working like one share. If the shareholder had less shares than one unit, it only had limited rights (mainly right to dividend). The main reason for this unit-system was to reduce the influence of sokaiya, gangster-like professional disrupters of shareholders’ meetings. “[T]he 2001 reform replaced the unit share scheme with another one that bundles shares in connection with voting rights. On the other hand, minority shareholders’ rights, including proposal rights, are now granted to such shareholders as have at least a certain percentage or number of voting rights.” (Tatsuta 2005, p.195) 46 The right to inspect the books was amended to 10% after structural impediment talks in the late 1980s. (Tatsuta 2005) The threshold ownership after the reform in 1950 was 3%. (Blakemore and Yazawa 1953, pp.20-21) 47 Prior to the law revision in 1974 a shareholder with at least 25% ownership could request an election of the board of directors according to cumulative voting, irrespective of the articles of incorporation. As of 1974 companies were allowed for the possibility “to discard this method, and nearly all corporations have such a provision in their charter” (Tatsuta 2005, p.194).

71 Chapter 3 Informal and formal institutions

company is not required to obtain shareholder approval for the merger. These procedures are applicable when the size of the target company is less than 5 % of the bidder company (Shishido 2004, p.14). Amendments of the commercial code, effective as of October 1999, permit Japanese companies to use stock swaps and stock transfers for corporate acquisitions and reorganizations. The stock swap and stock transfer can be used to create a holding company structure with wholly owned subsidiaries, as it makes it easier for parent companies to obtain 100% ownership in companies. With a stock transfer parent companies have the opportunity to use their own shares as currency to acquire (related or unrelated) companies. In a stock swap a parent company can exchange its own shares with those of a subsidiary company without approval of minority shareholders, if there is a special resolution by the shareholders’ meeting of the target company. Whereas prior to 1999 companies could only swap shares in mergers, it was now also possible in takeovers. This allowed cash constrained but highly valued listed companies to use their own stock to pay for takeovers. In 2001 a new statutory scheme became effective and provided parent companies a more flexible framework to separate business units through spin-offs and split-offs. With the corporate divesture system the shareholders’ meeting can pass a special authorizing resolution, and a company can “transfer an internal division to another company and be compensated with the stock of the acquiring company. Under this new type of business transfer, acquiring companies could be freed from needing to raise funds for the purchase” (Arikawa and Miyajima 2007, p.8). Before the amendment, assets to be transferred needed court supervised valuations and liabilities required individual approval from creditors (Milhaupt and West 2004, Ch.7). Important amendments in the commercial code were enacted in 2005: introduction of the short-form merger process, amendments to the simplified merger process of 1997 (effective as of May 2006), regulations regarding merger currency flexibility, and triangular mergers (effective as of May 2007). Simplified merger process - A vote of the bidder company’s shareholders is not necessary when the ratio of the merger consideration to the net assets of the bidder company’s assets is not higher than 20%. The previous consideration was a stock issue of no greater than 5% of the total outstanding stock of the bidder company. Short-form merger process – This process can be used when a bidder company merges with a subsidiary company of which it owns 90% or more of the voting shares. In this case approval of the target (subsidiary) company’s shareholders is not required. This speeds up the merger process and can be used to squeeze out minority shareholders of the target company.48 Merger currency flexibility - The new company law allows cash and other assets (including for example, stock of a foreign parent company) to be considered as payment in a merger. Previously, the merger currency was largely limited to shares of the bidder company and cash could only be used in combination with shares (Kodate 2005). Kodate (2005) further explains that according to the current law “[C]ash and the shares of corporations other than the surviving entity may also be used as consideration for a merger

48 The minority shareholders, on the other hand, will have the right to block the merger if it violates the law or the Articles of Incorporation, or if the terms are extremely unreasonable and threaten to have an adverse affect on the target company’s shareholders.

72 Section 3-3 Formal institutions related to M&A

if the requirements of the Industrial Revitalization Act, such as receiving approval of the relevant Japanese Ministers, are met” (p.3).

3-3-2 Antimonopoly Act The Antimonopoly Act aims to secure free and fair competition by prohibiting monopolies and anti-competitive or unfair arrangements. Mergers are not allowed if they restrict free competition in a particular industry, or if they are achieved through unfair trade practices. In order to control and prevent these situations, merging companies need to report their proposal to the Fair Trade Commission. The Antimonopoly Act was amended in 1998 “to reduce the scope of reporting and notification requirements regarding mergers and stockholding and to improve examination procedures” (JFTC 1998). As of January 1999 it was no longer necessary to notify each merger to the Japan Fair Trade Commission. The notification requirement is limited to mergers between a company with total assets of over ten billion yen (including the company’s parent and subsidiary companies) and a company with total assets of more than one billion yen (including the company’s parent and subsidiary companies). Also, notification is not required for mergers between (i) a parent company and its consolidated subsidiaries, and (ii) mergers between sister companies (when a parent company owns over 50% of a subsidiary’s stock).

Holding company - Effective as of December 1997, the Antimonopoly Act was amended in order to authorize the establishment of pure holding companies: this allows companies to change business segments into joint-stock companies, making it easier to buy and sell separate/individual segments. Also, within the holding company structure it is possible for absorbed companies to survive as separate entities. The system prevents friction in personnel and organizational matters in M&A deals. The pure holding company was banned in 1947 by the Antimonopoly Act. Only the operating holding company was allowed. A holding company possesses shares in subsidiaries and effectively controls these companies. Stock ownership of subsidiaries is 50% or more, either directly or indirectly. In effectively controlled companies the holding company is the largest shareholder and has stock ownership between 25% and 50%, including stock ownership through other subsidiaries. The pure holding company only has stock ownership and control of other companies and no business activities. The pure holding company’s main purpose is to control the companies in the holding and earn profits through dividend payment by these companies. The operating holding company has its own profit-seeking business activities as well as stock-ownership in other companies. In pre-war Japan pure holding companies had been important for the organizational and governance structure of Japanese companies (zaibatsu). They were banned because of their high involvement in the war-industry and it was argued that the pure holding company would not be in line with the aim to create a fair and free market, as stock ownership would not be flexible.

Financial institution stockholdings - After the war financial institutions were allowed to hold only 5% of stock from outside companies. This percentage was increased to 10% in 1953 but the 1977 reform reduced the maximum share ownership from 10% to 5% as of 1987. Of the financial institutions only insurance companies are allowed to hold 10%.

73 Chapter 3 Informal and formal institutions

3-3-3 Securities Exchange Law The most important regulations with reference to M&A in the Securities Exchange Law are related to a tender offer bid or takeover bid (TOB). A tender offer (“Tender Offer”) is a bidder company’s offer to a large number of unspecified persons to tender a company’s shares. Tender offer regulations function as a means to prevent that minority shareholders cannot participate in a specific transaction. Prior to 1971 tender offers were unregulated in Japan, and from 1971 until 1990 a ten-day waiting period and prior review of all offers by the Ministry of Finance were introduced. In 1991 this prior notification procedure was abolished and, instead, notification can be given at the start of the tender offer process. This will be the same day the notice of the bid is published in newspapers. It will still be necessary for a foreign offeror to give notice through the offices of a resident filing agent, but this no longer has to be a securities company or a commercial bank and can be a resident of Japan such as a lawyer or an accountant. The validity period of tender offers has been extended to 60 days (minimum period of 20 days). In 1991 the compulsory tender offer system was introduced as well (Milhaupt and West 2001, p. 19). The tender offer regulations need to be adhered to when (i) more than 5% (including shares held before the purchase) of a company’s outstanding voting rights is purchased off-market from more than 10 sellers during a 60-day period, and (ii) more than one third (including shares held before purchase) of the outstanding voting rights of a company from any number of sellers is purchased on and off-market within 60 days. In 2005 the regulations were amended in response to the use of the off-hour trading system by Livedoor. The regulations now also include purchases of more than one third of the outstanding voting rights through specific transactions (for example, off-hour trading systems such as Tokyo Stock Exchange Trading Network System – ToSTNet). Another important amendment was introduced after the purchase of Origin Toshu’s shares by Don Quijote in 2006. TOB regulations now also apply to serial purchases (other than acquisitions of newly issued shares), if these purchases result in the bidder company’s ownership to exceed one third of a company’s outstanding voting rights and the purchase (i) is made within a period of three months, (ii) exceeds 10% of outstanding voting rights, irrespective of the purchase method (including market-based acquisitions or acquisitions of newly issued shares), and (iii) is made off-exchange or through an off-hour trading system and exceeds 5% of outstanding voting rights (excluding an acquisition over a tender offer).

3-3-4 Other laws related to M&A Industrial Revitalization Law - The Industrial Revitalization Law (sangyǀ saisei hǀ) came into effect in October 1999. “The purpose of the Law is to systematically improve the productive potential of the management resources available to Japanese companies” (Jetro 2001). The industrial revitalization measures are three-fold: (i) facilitating business restructuring (special measures for companies with ministerial approval of restructuring plan), (ii) support for business start-ups and growth of SMEs, and (iii) stimulating R&D and technology development. The first measure, facilitating business restructuring, provides companies with special measures (for example, tax preferences and fiscal and financial support) when ministerial approval is received for the restructuring plan. These restructuring plans can be disposition of excess facilities but also M&A activities or other forms of restructuring. The law was amended in 2003, and the support measures for corporate restructuring were increased, as a means to cope with structural oversupply and excess debts. “More specifically, support has

74 Section 3-4 Summary

also become available with respect to projects including “management resource reuse projects” that are aimed to improve productivity of businesses transferred from other companies, the scope of simplified corporate restructuring has been expanded, and mergers for money or other companies’ stocks have become allowable” (Jetro 2001).

Civil Reorganization Law - The Civil Reorganization Law (minji saisei hǀ) came into effect on April 1 2000. The main goal of the reorganization law is to provide viable companies that experience difficulties with mechanisms to reorganize business. Reorganization can be in the form of restructuring debt, replacing management and/or reorganizing the corporate structure. Reorganizing the corporate structure can be achieved through mergers, acquisitions, or other M&A or restructuring activities. The law offers provisions for rapid execution of organizational procedures.

Accounting Standards - From fiscal year 1999 new accounting standards were implemented to improve international comparison. New regulations affecting M&A activity were the obligation to disclose consolidated rather than parent-only earnings in financial reports, and the requirement of reporting financial assets at market value instead of book value. Other regulations included introduction of statement of cash flows, introduction of tax effect accounting and market valuation of retirement benefits. Market value accounting was introduced for pension liabilities in 2001 and for holding companies in 2002 (Hattori 2004, Chapter 2).

3-4 Summary This chapter discussed the informal and formal institutions in pre-war and post-war Japan. We showed that informal institutions originate from Shinto, Confucianism, and Buddhism. Based on the principles of the latter two, Shǀtoku Taishi wrote the Constitution of Seventeen Articles in 604AD. This moral constitution stresses virtues, of which most are still present in contemporary Japan. Examples of the virtues are harmony, tolerance, modesty, trust, and importance of the group. Other important informal social institutions in Japan are the ie system and the dǀzoku system, which represent the relationships between individuals and companies in a group. Section 3-2 described the political development and its impact on the financial markets in the pre-war and post-war periods. In 1872 the National Bank Act was implemented to create a modern banking system, and in 1878 the Stock Exchange Act came into force. The Meiji constitution was promulgated in 1889 and changed the political system into a centralized bureaucratic state. Japan was placed under the command of SCAP after WWII, and SCAP’s policies were aimed at demilitarizing and democratizing Japan. Japan changed into a democratic country and a regulatory environment for a safe and stable financial system was set up. The stock market and the bond market are characterized by continuous deregulation in the 1980s and 1990s. Section 3-3 discussed the formal institutions related to ownership and M&A in the pre- war and post-war periods. Before WWII, unlimited authority in all financial and managerial matters was placed in the general shareholders’ meeting. After the war, minority shareholders gained a lot of power, and authority on managerial matters was transferred to the board of directors. The Antimonopoly Act banned establishment of pure holding companies and prohibited shareholdings between companies of the same pre-war

75 Chapter 3 Informal and formal institutions

corporate group. The latter prohibition was abandoned after the peace treaty was signed in 1952, and establishment of a pure holding company is authorized since 1997. Laws directly related to M&A were implemented from the late 1990s onward, with simplified merger procedures, merger currency flexibility, and easier notification procedures for tender offers. As of 2001 it is possible for companies to buy back their own shares and cancel or hold them as treasury stock.

76 4

Ownership and corporate governance

This chapter investigates the influence of informal and formal institutions on the financing decision of Japanese companies and the formation of corporate groups. As we explained in chapter 1, the financing decision results in the ownership structure that consequently leads to the corporate governance structure. The ownership structure and the corporate governance structure, constrained by the informal and formal institutions, subsequently have an important influence on M&A activity of companies. In our analysis the historical and context-specific background is important and we therefore use the Meiji period (1868-1912) as a starting point. Taking WWII as our breaking point, we look into the pre-war and post-war periods. In section 4-1 we start with a brief economic overview of the two periods. Section 4-2 investigates the financing decision, ownership structure, governance structure, and formation of groups in the pre- war period, and section 4-3 examines these for the post-war period. Section 4-4 summarizes and concludes the chapter.

77 Chapter 4 Ownership and corporate governance

4-1 Economic characteristics49 R The government played an important role in the initial stage of pre-war economic development, as traders had not accumulated sufficient wealth to make the necessary investments. R During the post-war period, government involvement was in the form of administrative guidance of companies (gyǀsei shidǀ). R In the pre-war period as well as the post-war period, companies in the same industrial sector worked closely together in trade associations.

This section provides the context-specific background for interpretation of Japanese companies’ financing decision and ownership structure. We start by examining the economic development during the pre-war period and the post-war period; in our discussion we pay special attention to the role of the government and trade associations. In sub-section 4-1-2 we look into administrative guidance and cartels during the post-war period.

4-1-1 Economic development As discussed in section 3-2, in order to adjust the treaties to equality, the Meiji government focused on the creation of a modern legal system and a capitalistic society to attain economic growth. As to the latter, the government took the initiative to establish companies equipped with Western machinery50, set up technical schools, and send students for studying and training purposes to Europe and the U.S. Foreign instructors, advisers, and engineers were also invited to come to Japan to help running companies and train technicians (Beasley 1963). The Kobushǀ (Ministry of Engineering and Public Works) was created in 1870 and ran companies ranging from lighthouses, telegraph lines, railroads, mines, shipyards, iron and steel mills, to model factories for manufacturing machines, glass, cement, and bricks. In 1880 government-owned industrial undertakings and properties totaled 3 shipbuilding yards, 51 merchant ships, 5 munitions works, 52 other factories, 10 mines, 75 miles of railways, and a telegraph system linking the most important cities (Allen 1946, p.30). Johnson (1982) argues that the unequal treaties resulted in the direct involvement of the government, because Japan could not protect private initiative in developing industries; tariff autonomy was not realized until 1911 (p.25). Beasley (1963) explains that the wealth of the Tokugawa merchants was too small to engage in activities that needed long-term and expensive investments. Their restricted activities, in particular regarding foreign trade, had not resulted in the required accumulation of wealth. During the 1880s, many of the government-owned companies were sold to private entrepreneurs. The government faced an increasing debt burden due to the deflationary policy and, in her desire to establish a modern economy, wanted companies to be managed by private initiative. Companies were primarily sold to entrepreneurs that had close

49 See: Pierce (1993), Rixtel (1997) and Suzuki (1989) 50 Allen (1946) indicates that enterprising daimyǀ and the shǀgun had already started to set up Western-style factories during the first half of the century. “The lord of Satsuma, for example, had founded works for the manufacture of pottery, cannon and cotton yarn. Indeed, the first cotton spinning mill in Japan was started in 1861 by this daimyo who imported Lancashire textile machinery for the purpose. The Shogun himself had founded Western enterprises, including two shipbuilding yards, during the later fifties” (p.29).

78 Section 4-1 Economic characteristics

connections with the government, the so-called (tokken) seishǀ.Theseishǀ were able to adopt new technologies and developed into leaders of zaibatsu groups. In line with the Meiji government’s policy, the seishǀ were committed to Japan’s economic development and to strengthening her military power. Examples of sales of government-owned companies are the Nagasaki Shipyard, Sado Gold Mine and Ikuno Silver Mine to Mitsubishi, and Miike Coal Mine and Shinmachi Silk Spinning Mill to Mitsui (Johnson 1982, p.23). The cost to the zaibatsu companies were often lower than the investments made by the government. Morck and Nakamura (2003) mention that “many state-owned enterprises were in dismal shape, and […] many privatized enterprises encountered serious difficulties” (p.26). According to Fukasaku (1992), the unsuccessfulness of companies was due to bureaucratic inefficiency and inexperience, and “the major significance of government enterprises lies in the fact that experiences in modern industrial entrepreneurship had been gained, and certain practices in going about it” (pp.20-21). Allen (1946) argues that “[I]t can be said with truth that there was scarcely any important Japanese industry of the Western type during the later decades of the nineteenth century which did not owe its establishment to State initiative” (p.30). The establishment of the government-led companies had laid the foundations for the economy, and their sale resulted in Japan’s economic growth. After the 1880s, the role of the government evolved into providing the economic environment in which domestic companies could grow and become more competitive vis- à-vis foreign companies. The government facilitated the required foundations for economic development by expanding the transportation infrastructure (railways and ports) and communication systems (telephone and postal services). Japan’s industrial take-off in the period 1885-1905 was driven predominantly by textiles, trade, and heavy industries such as iron, steel, and shipbuilding. Growth in these industries was spurred by the Chinese- Japanese War in 1894–95, the occupation of Taiwan, and the Russian-Japanese War of 1904-05. This period is also characterized by the development of the dual structure (nijǌ kǀzǀ) in the Japanese economy. There was a high degree of market concentration in leading industries, dominated by a small group of very large companies, supported by a large number of small- and medium-sized companies and affiliated suppliers. During the First World War (“WWI”) European companies stopped exporting, and this enabled Japanese companies to penetrate new markets such as India and South East Asia. Exports to existing foreign markets, China and the United States, also increased (Beasley 1963, p.214). Total exports boosted from 1.2 billion yen in 1914 to 4.5 billion yen in 1919, and GNP grew at an average of 6% in the period 1885-1915. Established zaibatsu became very rich in this period; paid-in capital of Mitsui Trading Company increased from 20 million yen in February 1918 to 100 million yen in 1919. A lot of new (holding) companies, Taishǀ zaibatsu, were also established by the so-called sensǀ narikin (wartime nouveaux riches). One of the most famous new zaibatsu was set up by Kaneko Naokichi and included companies such as Suzuki Trading, Kobe Steel, Harima Shipbuilding, Imperial Rayon (Teijin), Nihon Flour Mill, Great Japan Celluloid, and Hǀnen Refining (Johnson 1982, pp.89-90).

79 Chapter 4 Ownership and corporate governance

0.50

0.40

0.30

0.20 %

0.10

0.00

8 7 6 5 4 3 88 89 90 91 92 93 1885 1 1891 1894 1 1900 1903 1 1909 1912 1 1918 1921 1 1927 1930 1 1936 1939 -0.10

-0.20

Figure 4-1 GNP growth in the period 1886-1942 Notes Annual GNP growth calculated based on gross national expenditure as estimated by Ohkawa, Takamatsu, Yamamoto. Source International Financial Statistics Yearbook.

The WWI boom collapsed in 1920 and Japan experienced a severe financial recession as a result of increased competition from Western companies. As the quality of Japanese products was still inferior to Western products, total exports fell from 4.5 billion yen in 1919 to 2 billion yen in 1920. The first financial crisis occurred when the stock market crashed, and the index fell from 254.1 in February 1920 to 112.6 in September 1920.51 The so-called Ishii panic occurred in 1922 (a number of banks collapsed), and the Kantǀ earthquake of 1923 caused severe destruction in Tokyo and the Kantǀ region. The financial panic in 1927, due to financial problems of several banks, led to bank runs and closure of almost 40 banks. The old zaibatsu companies had greater and more diversified financial resources than other companies, and the effect of the economic downturn in the 1920s was therefore only limited. In contrast, the impact on zaibatsu companies established during WWI was strong and several went bankrupt (Johnson 1982, p.96). Although GNP growth fell considerably compared to 1915-19, it remained positive at an average of 1% over the period 1920-29. The lifting of the gold embargo in 1930, combined with the effects of the world depression of 1929, and the strict deflationary policy, hit the Japanese economy hard; GNP fell by 10% during the Showa depression of 1930-1931. The effects of the Showa depression were felt in all industrial sectors, but especially in the agricultural sector. Poverty, combined with dissatisfaction with the government, led to the emergence of military and rightwing groups in the beginning of the 1930s (Allen 1946). In the period from 1932 to 1945 Japan recovered economically from the 1920s and turned into a military state. The Manchurian Incident in 1931, when the military overtook Manchuria without government consent, indicates that the government could not control

51 The stock exchange index in 1913 at 100.

80 Section 4-1 Economic characteristics

the military anymore. The military tried to increase their influence further by undertaking two military coups in 1931 (both failed), and by assassinating the prime minister in 1932. Other groups that criticized politicians and the large zaibatsu companies also emerged. One of these groups is the Ketsumeidan (Blood Brotherhood Club) that murdered a former finance minister and the founder of the Mitsui zaibatsu. The military gained full political control 52 in 1937 and started the China-Japan war. Allocation of financial resources according to the market mechanism was replaced with planning by government and bureaucracy during the wartime period 1937-45. The government obtained authority to allocate funds directly to munitions and other war-related industries. Banks were used as intermediaries and therefore played an important role in companies' capital funding in this period (Kaizuka and Ono 1986). In these years, a designated bank system was implemented and large companies were assigned to one bank from which it should procure funding. Of the 2,240 companies in this system, 1,582 (more than 70%) were assigned to one of the five major zaibatsu banks.53 These zaibatsu banks belonged to zaibatsu groups that played an important role in the war-related industry. The zaibatsu banks' provision of capital to companies during the war-period was indispensable and in 1945 total deposits and credits had amounted to 45.7% and 67.2%, respectively (Aoki 1994).

Trade association - In addition to the government and the zaibatsu groups, the trade association was an important element in the economic development of pre-war Japan. The trade association can be defined as “a combination (or federation of combinations) of two or more firms that has as one of its principal purposes the furtherance of common business interests” (Schaede 2000, p.30). The associations in the pre-war period were aimed at sharing technical information, protecting members’ privileges, establishing trade rules, and endorsing credit for members. Later, the associations also controlled prices and eliminated the common practice of companies' raiding competitors for employees (Schaede 2000, p.240). The first trade association was the Association of National Banks that was formed in 1876 by the roughly 150 banks that were established since 1868. In industry, the first trade association was the Japan Paper Manufacturing Association in 1880, followed by the Japan Cotton Spinners’ Association in 1882. Trade associations were also set up in the linen yarn, home-produced petroleum, and regional coal mining industries (Lynn and McKeown 1988). It is important to distinguish the trade association from a cartel, as the functions of the latter are primarily limited to control of output and prices. Teranishi (2005) mentions that cartel agreements with these objectives were exceptional in the pre-war period. He explains that there were only 84 cartels in 1932 and that “7 were formed before 1914, 12 during 1914-26, and more than half (48) during 1930-32 after the shock of the return to the gold standard and enactment of the Controlling Important Industries Act (jǌgyǀ sangyǀ tǀsei hǀ)” (p.100). According to Caves and Uekusa (1976), the financial difficulties in the period 1920-31 resulted in a more important role for cartels in the 1920s and their rapid increase in the 1930s. In 1925 a law was passed that authorized cartels and provided for compulsory adherence by members. The Controlling Important Industries Act of 1931 gave the government the power to enforce adherence to the cartel at the request of more

52 During the coup three ministers were assassinated. After the military coup failed, Emperor Showa demanded the military to stop. 53 Dai-Ichi, Mitsui, Mitsubishi, Sumitomo and Yasuda.

81 Chapter 4 Ownership and corporate governance

than two-thirds of the companies in the industry. The government was authorized to set up compulsory cartels with the wartime controls of 1938, and had initiated and supervised 1,538 compulsory cartels by the time the war ended (Caves and Uekusa 1976). Schaede (2000) argues that cartels were only effective in recession and depression years. In economic downturns the members colluded, with the cartel functioning as a safety net, while in booming markets the members competed fiercely. Although the number of cartels was low and only increased during the war, already by 1886 there were at least 1,579 trade associations.

25.0

20.0

15.0

% 10.0

5.0

0.0

5 1 4 0 3 9 2 8 7 95 96 96 97 97 97 98 98 991 99 000 1952 1 1958 1 1 1967 1 1 1976 1 1 1985 1 1 1994 1 2 2003

-5.0 GDP Growth Inflation

Figure 4-2 GDP growth and inflation rate in the period 1952-2004 Source International Financial Statistics Yearbook (various issues).

Post-war economic development - Japan faced a great challenge after the war because the air raids had destroyed most of the production facilities, and capital and raw materials were scarce. An investment- and export-led strategy was implemented in order to obtain an inflow of raw materials and foreign capital. Japan was committed to reconstruct the economy and this objective was achieved in the mid-1950s when the take-off period of the Japanese economy started. The years from mid-1950s until mid-1970s are characterized by a high growth economy (double-digit GNP growth) and a balanced government budget. The high economic growth was supported by a high level of capital investment in the private sector, generated by high personal savings. Other important aspects were imported foreign technology and the large transfer of the working population from primary to secondary industrial sectors. 54 In 1964 Japan accepted article 14 of GATT (no governmental subsidies of exports), and article 8 of IMF. The latter meant Japan would also adhere to article 11 of GATT (no trade controls because of balance of payments

54 Overall, the supply of high-quality labor was supported by fast growing population.

82 Section 4-1 Economic characteristics

deficits). The same year Japan became a member55 of the Organization for Economic Cooperation and Development (OECD), and committed itself to trade liberalization and the removal of controls on capital transactions (Johnson 1982, p.263). Around 1973 the high-growth period ended with the shift to floating exchange rates, the first oil crisis and the rise in worldwide inflation rates. Figure 4-2 depicts the Japanese GDP-growth and inflation rate for the period 1952-2004. A period of lower growth commenced after 1973 and the general economic strategy shifted its focus to domestic demand. It also became apparent that the way the financial system had functioned would have to change. As explained in section 3-2, regulatory reform of the Japanese financial system became an ongoing process in the 1970s. During the 1980s GDP grew at an average of 6% and inflation was low. A lot of investments were undertaken in plant and equipment, resulting in a stock market and land price bubble. These bubbles burst in respectively 1989 and 1991, and companies had difficulty repaying their debts. This subsequently resulted in bankruptcies of financial institutions in the latter 1990s, and economic growth fell to an average of 0.89% over the period 1992-2004. Negative growth was experienced for the first time in 1998, and the Japanese economy stagnated during 1998-2004 at an average of -0.43%. Inflation was low in the first half of the 1990s (1.34%) and turned into deflation in the latter half of the 1990s and the 2000s.

4-1-2 Administrative guidance (gyǀsei shidǀ) Administrative guidance refers to the role of the government and bureaucracy in the economic development of post-war Japan. In the 1950s the government implemented policies to protect and foster domestic industries in order to achieve economic independence and full employment. During the 1960s this policy was changed to establishing solid economic growth and a higher living standard (income-doubling plan). In the latter half of the 1960s improvement of Japan’s international position was most important. The post-war economic plans of the government had three characteristics: they (i) indicated the direction of economic and social development, (ii) showed the policy direction of the government, and (iii) specified behavioural guidelines for people and businesses (Nakamura 1994, p.89). Table 4-1 shows the objectives and goals of the policy plans from the 1950s to the 1970s. The government guided companies and industrial sectors in various ways. When an industrial sector faced a business downturn, “a helping hand would be extended by the government in the form of cartel assistance, tax reductions or exemptions, and industry wide plant and equipment capacity expansion agreements (setsubi chǀsei)” (Nakamura 1994, p.66). Overall, the government provided companies with a safe economic environment in which plant and equipment investments were promoted. Government policies were significant in promotion and explanation of the industrialization process, but the role of the Ministry of International Trade and Industry (“MITI”) was most important in the implementation. Lynn and McKeown (1988) mention in this respect that “Japan is a parliamentary democracy with a parliament, the Diet, that is frequently characterized as reigning rather than ruling. It is the bureaucracy that rules” (p.91). Johnson (1982) argues that “[I]n general, prewar ministers had more influence over

55 “some seventeen temporary reservations to the OECD's code of behavior for members but "fully opened economy" was at last on the nation's agenda” (Johnson 1982 p.263).

83 Chapter 4 Ownership and corporate governance

their ministries than postwar ministers, a change that again reflects the rise in bureaucratic power in the postwar era” (p.55). In the implementation, MITI’s goals were the (i) promotion of resources toward favoured industrial sectors, (ii) promotion of larger operations in the form of large-scale plants in favoured industrial sectors, (iii) encouragement of import and dissemination of new technology, and (iv) protection of domestic markets from foreign products and companies, (v) preventing excessive competition in periods of lower growth, and (vi) discouragement or (occasionally) promotion of entry in industrial sectors (Caves and Uekusa 1976, p.149).

Table 4-1 Government policy plans for the period 1955-1971 Name of plan Five-year Economic Self-support Plan Planning period FY1955-60 Objectives Economic independence; full employment Goals Achievementofeconomicindependence&fullemploymentonthebasisofastable economy

Name of plan New Long-range Economic Plan Planning period FY1958-62 Objectives Maximum growth, higher living standard, full employment Goals To seek a steady rise in the national standard of living and to approach full employment by continuing to achieve as high a growth rate as possible while maintaining economic stability. Name of plan National Income-Doubling Plan Planning period FY1961-70 Objectives Maximum growth, higher living standard, full employment Goals To advance toward an outstanding improvement in national living standard and full employment. For these purposes, seek to maximize stable growth of the economy. Name of plan Medium-term Economic Plan Planningperiod FY1964-68 Objectives Correction distortions Goals To correct distortions, i.e., to bring in line with the pace of economic & social development those sectors lagging behind in production or living standards, and to seek the harmonious development of the economy & society. To promote social development. Name of plan Economic & Social Development Plan Planning period FY1966-71 Objectives Development toward balanced & enriched economy & society Goals To create the basic conditions for securing the position of the Japanese economy in changing international milieu and for people's enjoyment of a full life appropriate to that position; and to develop toward a balanced & enriched economy & society. Source Summary of Nakamura (1981), Table 39, pp.84-88.

Odagiri (1992) divides the role of MITI in the post-war period up to the first oil crisis into three periods. The first period starts in 1945 and ends before 1950, when priority production (keisha seisan) was given to the industrial sectors coal-mining and steel. Providing the coal-mining industry subsidies, low interest loans and so on, resulted in low cost coal that was necessary for the steel industry. The 1950s, the second period, is characterized by industrial rationalization (sangyǀ gorika) and involved promotion and introduction of new equipment and technology. Infant industries, such as the automotive, electric and electronic equipment, and petrochemical industries (p.290), were supported and protected from foreign competitors with quotas and tariffs (ikusei). The third period,

84 Section 4-1 Economic characteristics

the 1960s, can be termed as the new industrial order (shin sangyǀ taisei). Trade and foreign exchange were liberalized, and import restrictions decreased from 466 in 1962 to 100 in 1972, and 27 by 1975 (Odagiri 1992, p.288). In the 1960s MITI felt that Japanese companies were not large enough to be competitive vis-à-vis foreign companies and mergers and investments were encouraged. After the first oil crisis government policy was directed towards the promotion of knowledge-intensive industries and adjustment of structurally depressed industries. Regarding the latter, “[T]he government used the Temporary Measures Law for the Stabilization of Designated Depressed Industries […] to legalize the formation of cartels in order to co-ordinate scrapping production facilities in such industries as aluminium, synthetic fibres, and petrochemicals, and to create a fund to guarantee the firms against any debts incurred” (Odagiri 1992 p.289). In the period after 1973 government policy was directed to the promotion of knowledge-intensive industries and adjustment for structurally depressed industries. In the 1980s MITI focused on encouraging imports and discouraging exports, by persuading the industrial companies to adopt voluntary export restraints.

Cartels - As mentioned above, cartel arrangements between companies were an important element of administrative guidance in the post-war period. The Japan Fair Trade Commission (‘JFTC’) controls the formation of cartels based on the Antimonopoly Act, which prohibits unreasonable restraints of trade. Only two types of cartel agreements are permitted with JFTC consent: the depression cartel and the rationalization cartel. In a depression cartel the producers decide to limit production due to an excess of supply over lacking demand. This cartel is aimed at preventing failure of companies in an economic downturn when prices fall below average cost. The rationalization cartel is aimed at colluding on aspects such as the industry-wide exchange or restriction of technology, technical improvement, product standardization and quality improvement, cost reduction, or making collective use of facilities (e.g. transportation or storage) (Caves and Uekusa 1976, p.143). Although these two cartels were permitted according to the Antimonopoly Act, because it was necessary to get JFTC approval, they were rarely used by MITI. In the period from 1953 to 1984, JFTC approved seventy-one depression cartels and thirteen rationalization cartels (Lynn and McKeown 1988, pp.41-43). More important for MITI’s administrative guidance was the ‘approved association’, which only needs an approval license from the regulating ministry (Schaede 2000). Hadley (1970) refers to the ‘approved association’ as “Administrative Guidance Cartels” because MITI recommended companies in the association to adhere to its guidance (p.381). Companies were inclined to do so, because MITI had various tools to support its recommendations, such as easy access to capital, tax breaks, approval of licenses, authorization of foreign technology import, and permission to establish joint ventures. If companies adhered to the MITI policy these activities were easy and quick, but if they did not the process could be cumbersome, long and costly56 (Johnson 1982; Caves and Uekusa 1976). Schaede (2000) indicates that in 1996 there were 2,223

56 Caves and Uekusa (1976) describe how MITI handled a situation in which Sumitomo Steel did not follow the prescribed policy. “Sumitomo, seeking to break into the circle of leading steel makers and equipped with a World Bank loan, constructed a large integrated mill in 1959 that greatly increased its capacity. When a 1964 recession led the steel makers to negotiate output quotas among themselves, Sumitomo initially refused to go along with the low quota dealt to it. The company's president was summoned to the ministry, however, and wisdom prevailed” (p.55).

85 Chapter 4 Ownership and corporate governance

approved associations. In the same year there were 9,655 voluntary associations, which are self-regulating trade associations independent of a regulating agency.

4-2 Pre-war financing and ownership R The most important financing method in pre-war Japan was internal financing. R Equity financing started to decline gradually from 1942 onwards, instead of the latter 1930s as commonly claimed in previous research. R A zaibatsu group is a conglomerate of companies, organized in pyramid form, owned by rich families and/or a family-owned holding company. R Pre-war corporate governance was concentrated within the holding company of the zaibatsu group.

It is commonly argued that, in comparison with the post-war period, Japanese companies relied more on equity finance than bank loans in the pre-war period. Okazaki (1999a) mentions that “pre-war Japanese firms had built-in systems for shareholders to monitor management. This was an important condition that allowed stocks and shares to be the main channel for capital provision” (p.106). Hoshi and Kashap (2001) argue that “if one compares the prewar system to the postwar US system and postwar Japanese system, the US system has more in common with prewar Japan” (p.50). We examine the importance of bank loans and equity finance with aggregate data in this section and in the next chapter we investigate the financing decision and ownership structure of Mitsubishi zaibatsu companies. Before we discuss the external financing decision, it is important to realize that, in accordance with the pecking order of financing, internal financing was the most important source of funding in the pre-war period. We show this with detailed data in our case study of the Mitsubishi zaibatsu in the next chapter, and Teranishi (1994) mentions that “[I]mportant industries outside the reign of the zaibatsu, such as cotton spinning and railroads, also used internal financing” (p.53). Table 4-2 shows that internal financing was most important in total funding up to 1936 (51.6% in 1935 and 47.4% in 1936). After 1937 the importance of external financing increased and, when looking at external funding in relative percentages, new share issues were more important than loans in the years 1934- 38.

86 Section 4-2 Pre-war financing and ownership

Table 4-2 Sources of funds for companies in the period 1931-1944 Internal Loans Shares Bonds 1931 43.4 7.7 16.9 32.0 1932 181.8 -88.6 -26.5 33.3 1933 105.2 -32.4 -3.9 31.1 1934 54.8 -13.0 2.8 55.4 1935 51.6 14.4 1.0 32.9 1936 47.4 21.3 -2.3 33.5 1937 33.3 31.3 -0.1 35.5 1938 30.5 29.6 5.4 34.6 1939 27.1 40.5 7.9 24.5 1940 30.4 37.3 5.5 26.7 1941 33.6 27.2 10.1 29.1 1942 41.5 29.1 7.6 21.9 1943 30.3 39.2 7.8 22.6 1944 24.2 58.5 8.3 9.1 Notes Percentages calculated based on data collected from mentioned source and also its underlying sources. The underlying sources are indicated in the notes of figure 4-3 as they are used to indicate the flow of funds during the period 1903-1945. The percentages of table 4-2 are frequently used in the discussion on pre-war financing in Japan; Hoshi (1995) for the period 1934- 1944, and Hoshi and Kashap (2001) for the period 1931-1957. Hoshi and Kashap (2001) only show the “Sources of External Funds for Industries: 1931-57” (p.85), and therefore fail to present a full overview of the sources of funds. Source Based on data collected from statistical yearbook “Supply of Industrial Funds (Increase and Decrease (1931-1984)”

Looking at external finance as a percentage of GNP, rather than the mentioned relative percentages, provides additional insights. Figure 4-3 shows bank loan, equity, and bond financing as a percentage of GNP for the period 1904-44. The figure indicates that in 1906, after the Russo-Japan war, loan financing was most important at 11.2% relative to GNP (equity financing was 2.1%). The figure also unveils that, apart from 1906, bank loan and equity financing were not very important for Japanese companies prior to 1915. In the period 1904-15 the average increase in bank loans (excluding 1906) and equity funding was respectively 3.3% and 2.6% as percentage of GNP. These findings confirm that internal financing was the most important source of funding. Resulting from increased production during WWI, financing through loans and equity increased considerably. Loans started increasing in 1916 and equity in 1917 by respectively 11.4% and 10.0% of GNP, showing double-digit growth in the next 3 years. In the period 1916-19 the average increase in loans was 14.5%, and in the period 1917-20 the averge increase in equity financing was 11.3%. In the period 1922-27 the use of bank loans returned to the level prior to WWI at 1.7%. Bond financing increased in importance and averaged 2.7% in this period. Stock financing returned to the pre-WWI level as well, averaging 2.8% in the period 1924-27. Figure 4-3 shows that in this period the annual increase in stock financing was 2.9% and in bond financing 4.0%. After the third bank crisis in 1927 the loans financing turned negative at an annual average of -2.3% until 1935. The most important aspect in pre-war financing that the figure shows is Japanese companies’ financing as of 1935. It is commonly argued that the post-war main bank system started in 1937, because Japanese companies turned to loans in stead of equity financing. In the war-period capital-intensive industrialisation took root and companies needed to make a lot of new and expensive investments. These investments caused companies to run out of internal funds quickly and made them adjust their financing decision. Hoshi and Kashap (2001) argue that “bank financing was increasing steadily

87 Chapter 4 Ownership and corporate governance

during 1937-45, while the importance of stock financing and bond financing was declining gradually. By 1939 new equity financing had become much less important than bank borrowing” (pp.85-86). Okazaki (1999a) states about the period 1936-40 that “the shift to the wartime economy led to a reduced role for the capital markets, which before the war had been the main channel of funds to the corporate sector” (p.112). These conclusions are based on the percentages of “Supply of Industrial Funds” provided by the Bank of Japan, but these data do not illustrate the trend in financing. For example, in 1939 total funding was almost 7 billion yen, of which loans accounted for 55.6% and equity finance for 33.6%. It appears that equity had become much less important, but the total equity financing amount of 2.3 billion yen was the highest level in the pre-war period. Equity financing did not decrease in the period 1937-43, but had increased to almost 4 billion yen in 1943. Because the stock investments of financial institutions were 9 billion yen in this year, we can assume they played an important role in equity financing of companies. The average annual stock financing in the period 1937-43 was 2.9 billion yen and the average security investment of financial institutions was 4.6 billion yen.

0.25

0.20

Russo-Japan War 0.15

0.10 %

0.05

0.00

5 1 3 7 9 5 7 1 3 9 903 909 91 91 923 2 3 1 190 1907 1 1 191 1915 1 191 1921 1 19 192 1929 19 193 1935 1937 193 1941 1943 -0.05 First Bank Crisis Third Bank Crisis -0.10 Loan as % of GNP Bond as % of GNP Stock as % of GNP

Figure 4-3 Financing as percentage of GNP in the period 1903-1944 Notes We collected the annual amounts of outstanding loan, bond and stock capital for the period 1903-1944. From these data we extrapolated the annual increase or decrease in funding. For each funding source this latter amount is divided by total GNP in the respective year. The data used for the period from 1930 though 1944 are similar to the external funds in table 4-2. As mentioned earlier, our percentages are based on the underlying sources and we calculated the percentage of external funding from 1904-1931. Sources GNP data from 1904 through 1940 are estimates by Ohkawa, Takamatsu, Yamamoto and. from 1941 through 1944 from Economic Planning Agency “Annual Report on National Income”. Stock data are from the National Tax Administration Agency, Ministry of Finance, “Taxation Statistics”. Bank loan data from Bank of Japan and consists of loans and discounts of all financial institutions. Bond data from the Bond Underwriters Association and shows the difference between issues and redemption of industrial bonds.

It is unlikely, however, that the equity financing in the 1940s was driven by an economic rationale. The period 1940-45 is characterized by an increasing role of the government and banks in strategic and financing decisions. Various war-related regulations were

88 Section 4-2 Pre-war financing and ownership

implemented to reduce the influence of stockholders and to place managers under direct control of bureaucrats. One of the first regulatory measures related to companies’ financing decision was “the compulsory lending system (Meirei Yushi Seido)oftheIBJinMarch 1939 (Kaisha Rieki-Haito oyobi Shikin Unyo-rei). Lending and security subscriptions over certain amounts had to be approved by the government, and those not directly or indirectly related to expansion of productive capabilities of the munitions industry were not approved” (Teranishi 1994, p.65). Figure 4-3 shows that the various regulations made loans a very important financing method, and that equity financing started to decline gradually from 1942. The Designated Financial Institution System for Munition Company Financing that started in January 1944 had the largest effect on the financing decision. It resulted in loan financing to jump to 14.8 billion yen, and equity financing to drop to 1.7 billion yen.

4-2-1 Pre-war zaibatsu Prior to WWII, large companies in Japan were often organized in so-called zaibatsu groups. These groups are conglomerates of companies, organized in pyramid forms, owned by rich families and/or a family-owned holding company. The holding company controlled the principal operating companies of the group, and the latter controlled many other companies through subsidiaries and shareholdings. There were three zaibatsu types with each their own financing method. At the end of WWII the four largest zaibatsu groups (Mitsui, Mitsubishi, Sumitomo, and Yasuda) controlled about one quarter of the paid-in capital of all incorporated companies (Caves and Uekusa 1976, p.60). These zaibatsu belong to the first type, Old zaibatsu, and were the most important companies in the Meiji period. The Mitsui zaibatsu and Sumitomo zaibatsu originated in the Tokugawa period (1600-1868), and the Mitsubishi zaibatsu and Yasuda zaibatsu were established in the Meiji period. These zaibatsu had close relationships with politicians who supported these businesses to establish strong domestic industries (in stead of foreign companies). The groups’ main activities involved the industrial sectors trading and finance, they were very profitable and up to WWI “expanded their activities by reinvesting profits” (Teranishi 1994, p.53). The second zaibatsu type, the Taishǀ zaibatsu, emerged as a result of the export-led boom during WWI (1916-1920). Various merchant families had engaged in the manufacturing industry during the WWI boom and had become very rich. They wanted to expand their businesses while keeping control over the company. The main source of financing for these zaibatsu was bank loans from a so-called “organ bank”. Examples are the Asano, Kuhara, Okura and Suzuki groups. In the bank crisis of 1927 a lot of these organ banks fell into trouble, and this led to the collapse of most of these zaibatsu. A famous example is Suzuki Shoten with its ‘organ bank’, the Taiwan bank. The Shinkǀ zaibatsu, New zaibatsu, emerged in the 1930s when the heavy and chemical industries started to develop in Japan. These groups had close relations with the military authorities and took advantage of the military demands, especially the munitions boom in the 1930s. The groups were technology-oriented with main activities in mining, chemicals and manufacturing, and the business leader had an engineering background. Each group had relations with various financial institutions and also relied on the stock market for financing. The largest of these zaibatsu were the Nissan, Nichitsu, and Riken groups. Some of these zaibatsu had accumulated a lot of capital during WWI and focused on sustaining continuing growth. Other groups, upon initiative of the military, made investments in the

89 Chapter 4 Ownership and corporate governance

Japanese colonies. The old zaibatsu were reluctant to make investments in heavy and chemical industries, which provided the Shinkǀ zaibatsu an opportunity to grow in these fields (Caves and Uekusa 1976, p.60; Shiba and Shimotani 1997). Depending on the type of zaibatsu, the companies relied on internal, equity, or debt financing. Figure 4-3 showed an important increase in the issue of equity and use of bank loans during the years 1917-20, when Taishǀ zaibatsu emerged. In the 1930s the issue of equity increased as the Shinkǀ zaibatsu arose. Teranishi (2005) explains that the use of a financial intermediary was a common financing method for non-zaibatsu companies. “[M]erchants (the commercial sector) were the major borrowers from banks. […] These funds were used either to invest in equities of large corporate firms or to lend on to small and medium enterprises and indigenous producers, including farmers. Equities were often used as collateral for the loans” (p.61).

4-2-2 Pre-war corporate governance The financing decision of a company results in its ownership and corporate governance structure. Okazaki (1999a) shows that the holding company played an important role in corporate governance within the zaibatsu group. In his comparison between the board of directors of zaibatsu and non-zaibatsu companies, he finds that “monitoring through direct participation in management was awarded only a relatively small role in the zaibatsu firms. What substituted for this was the systematic monitoring by the holding company of all its subordinate firms” (pp105-106). Okazaki (2000) explains the importance of the holding company in the decision making process and lists the “Rules on relationship between the core affiliated companies and Mitsubishi Goshi” (1918) as follows: a. Directors and auditors of a core affiliated company should be registered at Mitsubishi Goshi. b. Rules enacted by Mitsubishi Goshi should apply to the core affiliated companies. In case a core affiliated company enacts different rules or important rules, approval by the president of Mitsubishi Goshi is necessary. All the rules enacted by the core affiliated companies should (be) reported to Mitsubishi Goshi. c. Budgets and settlements of a core affiliated company should be approved by the president of Mitsubishi Goshi. d. A core affiliated company should send the expected fund flow and expected profit and loss to Mitsubishi Goshi. e. The chief of the Section of Audit, Department of General Affairs, should audit the accounts of the core affiliated companies, according to the order of the president of Mitsubishi Goshi. f. The staffs of a core affiliated company should be screened and employed by Mitsubishi Goshi. The personal affairs of the stuffs above the counselor (sanji) of the other stuffs should be sent to the president of Mitsubishi Goshi, by way of the Section of Personal Affairs, the department of General Affairs, in advance.

Rules related to raising funds by core affiliated companies were also established. Okazaki (1999a) explains that “a core affiliated company should concentrate operational accounts to the Bank department of Mitsubishi Goshi. Mitsubishi Goshi could monitor the operations of a core affiliated company through the Bank Department” (p.9). According to Odagiri (1992), ownership was separated from managerial decisions and competent

90 Section 4-3 Post-war financing and ownership

managers could (sometimes with difficulty in convincing the holding company) pursue dynamic investment and diversification strategies that were very important for the growth of the zaibatsu (p.170). The Taishǀ zaibatsu that emerged during the boom of WWI were characterized by their high degree of bank financing from an organ bank. The bank did not diversify, leading to the situation that, if the company would get into trouble, the bank would fail as well. The Shinkǀ zaibatsu, established in the 1930s, were probably the only companies with strong similarities to U.S. companies. They used stock financing and had relations with various financial institutions. So, it is likely that for this type of zaibatsu the stock market was important in the corporate governance process. However, to compare these zaibatsu with the U.S. stock market appears to be difficult as they emerged in the 1930s and had important relations with the military.

4-3 Post-war financing and ownership R Individuals did not sell their shareholdings in the 1950s, but were unable to purchase new shares. R A vertical keiretsu is a group of companies that take part in the production process of a core company, or parent company. R A horizontal keiretsu, also referred to as kigyǀ shǌdan, is a non-hierarchical group of independent companies in various industrial sectors. R Post-war corporate governance is characterized by the main bank system: a special relationship of a company with a bank, the so-called main bank. R The deregulation of financial markets weakened main bank relationships, and the 1990s are characterized by unwinding of cross- and long-term shareholdings.

In this section we examine the financing decision of Japanese companies in the post-war period. We start our discussion by looking into the effect of the dissolution of the zaibatsu- groups, as described in sub-section 3-2-2, on companies’ ownership structure. While the stock market was closed in the period 1945-49, the government arranged the so-called democratization of securities, which involved massive stock transfers from zaibatsu families, holding companies, and main member companies to individual investors. Priority in purchasing government-owned stocks was given to companies' employees and residents of localities in which plants were located (Oka 1993).

91 Chapter 4 Ownership and corporate governance

1,200 80 70 1,000 60 800 50

600 % 40

Topix Index 30 400 20 200 10

0 0 1945 1949 1950 1951 1952 1953 50 51 52 53 54 55 56 57 58 59 19 19 19 19 19 19 19 19 19 19 Financial Securities company Company Individual

Figure 4-4a Nikkei 225 daily stock average price index Figure 4-4b Share ownership in 1945, 1949-53 Notes The percentages of figure 4-4b are calculated based on the total number of shares listed on the First section of all stock exchanges in Japan, which includes holdings of governmental and public institutions, investment trusts, other domestic companies, and foreigners. Source Figure 4-4a based on stock price data from the Development Bank of Japan. Percentages of figure 4-4b are calculated based on Nomura Securities “Shareholders of Listed Companies (The First Section of the Market)”. International Financial Statistics Yearbook.

Although holding companies were prohibited, the zaibatsu families had lost their controlling power, and top executives were purged, the loyalties among the former zaibatsu companies had survived. Co-ordination of the groups’ activities was restored in the form of "presidents' clubs", wherein all pre-war core companies and financial institutions participated, and the group’s bank played an important role (Caves and Uekusa 1976, p.63; Odagiri 1992, p.281). Miyajima (1994) explains that the management of zaibatsu companies tried to stabilize stockholdings in three ways. The first method was through promotion and financing of stockholding by employees. Second, by asking “individuals and securities companies outside the corporate structure to hold their companies’ stock through unofficial contracts” (p.321). Third, financial companies of other zaibatsu groups were asked to hold their shares in exchange for deposits. The result of the democratization process was that the percentage of individual holdings at the Tokyo Stock Exchange had increased to 69.1% in 1949, from 53.1% in 1945 (National Stock Exchange Council). Loss of Japan's production facilities during WWII meant that economic revival was slow and companies’ profitability low. The TSE re-opened in June 1949 and only 80 companies out of the 502 listed paid dividends (Oka 1993). Oka (1993) argues that because dividend payments were almost non-existent, individual stockowners started to sell their stocks to institutional investors such as banks. Since most stocks paid no dividend, individual investors thought they could better transfer their funds to banks for at least some financial margin in terms of deposit interest (during the period 1945-58 this was fixed at 5% (Akimoto 1993), and in period 1959-71 at 4%57). According to Miyajima (1994), the declining share prices induced individual shareholders to sell, and Teranishi (2005) explains that the buyers of those shares were mainly individuals that had made huge profits from the Korean War boom. This situation resulted in a high risk of hostile takeovers in Japan. In sub-section 7-1-1 we present previous research that argues that horizontal keiretsu were formed during this period in order to prevent hostile takeovers. In sub-section 7-2-2 we give our own

57 International Financial Statistics Yearbook

92 Section 4-3 Post-war financing and ownership

interpretation of the hostile takeovers and formation of horizontal keiretsu in this period according to our institutional model that will be introduced in chapter 6. The so-called yield revolution in the high growth period, when dividend yields fell below those of deposits and savings, strengthened the sale of stocks by individuals even more (Akimoto 1993). During the high growth period the individual shareholding percentage declined from 53.2% in 1955 to 32.9 in 1975. “[T]his was due to two effects, outright selling and a failure to buy new equity issued by companies” (Hoshi and Kashap 2001, p.97). It is generally argued that the stock market in the period from 1949 to 1975 was characterized by a transfer of stock ownership from individuals to other market participants. Figure 4-5a indicates stockownership in percentage of shares during the period 1950-75 for the following groups: individuals, financial institutions, and companies. We agree with Hoshi and Kashap (2001) that the percentage of individuals’ shareholding fell during the high-growth period. On the other hand, we do not find any evidence of falling share ownership by individuals. Figure 4-5b shows that, in line with an increasing number of total shares, the number of shares held by individuals increased as well. It is possible that employees of companies started selling their shares (Miyajima 1994), but it appears that the buyers were predominantly other individuals, as argued by Teranishi (2005).

70 70,000 60 60,000 50 50,000 40 40,000 % 30 30,000

20 Shares (1,000) 20,000 10 10,000 0 0

8 0 2 4 6 8 0 2 6 8 4 5 6 6 6 6 68 70 72 950 954 96 964 96 9 9 9 9 9 1 1952 1 1956 195 196 1 1 196 1 1970 1972 197 1950 1952 1954 1956 19 19 19 1 1 1 1 1 1974 Financial Company Individual Financial Company Individual

Figure 4-5a Stockownership 1950-75 (%) Figure 4-5b Stockownership 1950-75 (shares) Note Figures 4.5a and 4.5b are based on the total number of shares listed on the First section of all stock exchanges in Japan. The percentages in figure 4-5a are calculated based on the total number of shares, which includes holdings of governmental and public institutions, investment trusts, other domestic companies, and foreigners. Source Calculation based on Nomura Securities “Shareholders of Listed Companies (The First Section of the Market)”

Stock ownership by companies and financial institutions started to increase as of 1952, when the Peace Treaty was signed. Especially the holding of shares between former zaibatsu companies changed considerably. Miyajima (1994) shows that cross- shareholdings of 18 same-line companies of the former Mitsubishi zaibatsu jumped from 1.8% in 1951 to 10.8% in 1952. Another important increase in shareholdings by companies and financial institutions occurred during the latter half of the 1960s after the collapse of the stock market in 1965. This crisis was primarily caused by the bad performance of investment trust companies that started to sell a lot of their shares during the depression years of the late 1960s. If we return to the falling ownership percentage of individuals during the high-growth period, it seems to be partly explained by the failure of individuals to buy new shares. We have summarized data on security financing of companies at all stock exchanges in table 4-

93 Chapter 4 Ownership and corporate governance

3. The table shows that up to 1965 only 6% of the issued shares was offered to the public. An upward trend of public offering started in the 1970s as financial markets were being deregulated.

Table 4-3 New share issues in the period 1956-1990 1956-60 1961-65 1966-70 1971-75 1976-80 1981-85 1986-90 OfferingtoShareholder94948851241913 Public Offering 6 6 11 45 70 72 57 PrivatePlacement--115553 Exercise Warrants ------4 26 Total 100% 100% 100% 100% 100% 100% 100% Note Percentages calculated as average of annual percentage of respective period. Source Calculated based Nomura Securities “Sources of Increase in Outstanding Shares (All Stock Exchanges)”

4-3-1 Post-war financing decision Because Japanese companies had virtually no internal funds left after WWII, their reliance on external funds persisted. In chapter 3 we explained that it was difficult for Japanese companies to enter the stock market, and that raising capital through the issue of stock was highly unattractive because this had to be issued at face value. The collateral requirements and the high issue costs of corporate bonds were the reason these were not used during this period. In the immediate post-war period the stock market and bond market were therefore virtually non-existent. As a result, companies could only finance their operations through bank loans and this made the corporate sector in Japan the largest deficit unit of the Japanese economy in the period 1945-73. The high levels of loans, combined with share ownership by financial institutions, sometimes resulted in a special relationship between companies and a bank: the main bank system, as will be discussed in sub-section 4-3-3.

Table 4-4 Net sources of funds for Japanese non-financial companies 1962-64 1965-69 1970-74 1975-79 1980-84 1985-89 1991-94 1995-97 Internal 39 50 43 52 56 53 71 91 Loans474352423934236 Bonds 23182-2 11 6 Stock 334535 Total 97% 99% 100% 100% 100% 100% 100% 100% Notes Percentages are the average of the years in each sub-period. Sources Bank of Japan, Economic Statistics Annual; Ministry of Finance, Flow of Funds Tables (no distinction made between Stock and Bonds for period 1962-69).

Around 1973-74 the financing decision of Japanese companies started to change because of a diversity of influences. Most important were (i) the increase in the corporate liquidity after the high growth period, and (ii) the deregulation process that opened new fund-raising possibilities for Japanese companies. Table 4-4 shows the relative shares of funding sources by non-financial companies in Japan. From the mid-1970s onward, external fund- raising by Japanese companies declined as a result of the high internal reserves that had been built up during the high growth period. At that time companies also started to look for other ways to raise funds, which was possible as a result of the deregulation of the financial system. This move towards different financing methods is manifested in the increase of corporate stock- and bond issues.

94 Section 4-3 Post-war financing and ownership

The management of companies also started to look for cost-cutting possibilities with reference to the financing decision and began to raise funds on international capital markets. The amount of foreign bonds increased from 600 billion yen in 1975 to 3,600 billion yen in 1991, which raised the share of foreign bonds in securities issues from 3% to 25%. As table 4-4 shows, however, this is a very small percentage in total external funding that continued to be dominated by loans from private financial institutions. Indirect financing in the form of loans started to decline after the stock market and real estate bubbles burst. Loans by private financial institutions fell by 400 billion yen in 1994, and as external direct financing increased Japanese companies started to depend more on securities markets. In this process the main bank relationship also weakened because the companies looked for alternative ways to obtain and invest funds.

4-3-2 Post-war keiretsu Two important industrial groups can be distinguished in the post-war period, the vertical keiretsu and the horizontal keiretsu.

Vertical keiretsu - The vertical keiretsu group comprises companies that take part in the production process of a core company, or parent company. Companies in the group can be wholly or partially owned by the parent company (such as subsidiaries and affiliates) or independent companies. It is a structure of shareholdings, including technology and personnel transfers from the core company to first-tier suppliers, from first-tier to second- tier, and so on. The participating companies are assemblers and subcontractors, and part of a subcontracting system (Nishiguchi 1994; Odagiri 1992). Westney (2001) argues that “[T]he vertical group structure clearly has a strong country component: it is the dominant mode of business enterprise in large Japanese firms across industries” (p.157). The structure of a core company with a network of subsidiaries is present in all major industrial sectors such as heavy industries, construction, food and beverages, chemicals, energy, steel, electronics, automotives, trading companies, department stores, transportation and real estate. The vertical keiretsu also results in a difference in size of Japanese and U.S. large companies. Westney (2001) indicates that the number of employees of Japanese companies is significantly lower than U.S. companies on a non-consolidated basis. In the year 1987 the number of employees was only 65,000 for Toyota, compared to 765,000 employees of General Motors. Other examples are Canon with 15,000 employees versus Xerox with 113,000, and Fuji Photo with 11,000 employees versus Kodak with 145,300. Westney (2001) concludes about the comparison of companies in the two countries that “[C]learly a more appropriate target for benchmarking was the group rather than the lead firm of the group” (p.156).

Horizontal keiretsu - The horizontal keiretsu, also referred to as kigyǀ shǌdan,arenon- hierarchical groups of independent companies in various industrial sectors. The post-war kigyǀ shǌdan differ from the pre-war zaibatsu; they do not have a central decision-making unit and are not under tight financial control from a parent company. These keiretsu groups are formed around a city bank and/or general trading company (sǀgǀ shǀsha). Groups of companies that consist of a city bank are also referred to as financial keiretsu, kinyǌ keiretsu. In these groups the city bank functions as the main bank of the group companies. The group companies also have relations with other financial institutions such as insurance

95 Chapter 4 Ownership and corporate governance

companies and trust banks. The financial institutions provide member-companies debt financing, have a considerable ownership in member companies, and dispatch members to the board of directors. The kinyǌ keiretsu are organized around the six large city banks and because each bank also belonged to a kigyǀ shǌdan,thekinyǌ keiretsu overlaps the kigyǀ shǌdan (Odagiri 1992). The six horizontal keiretsu that were formed in post-war Japan are Mitsui, Mitsubishi, Sumitomo, Dai-Ichi Kangyou, Fuji, and Sanwa. Mitsui, Mitsubishi, and Sumitomo are usually called ex-zaibatsu groups as they find their origins in the respective zaibatsu. The Dai-Ichi Kangyou, Fuji, and Sanwa keiretsu, centered on respectively the Dai-Ichi Kangyo Bank (DKB), Fuji Bank and Sanwa Bank, are essentially kinyǌ keiretsu. An important characteristic of these groups was the “one-setism”, which meant that in each group there was only one company per industry in order to prevent intra-group competition (Caves and Uekusa, 1976). The “one-setism” did not result in a cartel, but in arrangements of “repeated oligopolies” (Hadley 1970). In each industry, the top three or four leading companies are/were from one of these horizontal keiretsu.

4-3-3 Post-war corporate governance The post-war Japanese corporate governance is characterized by the horizontal keiretsu and the main bank system. 58 Companies in a horizontal keiretsu have cross-shareholdings between the different stakeholders of the companies. Among the stakeholders are suppliers, customers, trading companies, other affiliated companies, and the main bank. The shareholdings are argued to be based on trust and selling of these shares is not done (violation of trust), which reduces the pressure to develop a stock market to trade (Jacobson and Aaker 1993; Sheard 1989; Kester 1991).The horizontal keiretsu has various functions, such as exchange of information (e.g. through monthly meetings of “presidents” clubs), the pursuit of growth through in-group joint ventures, and mutual insurance. As explained in section 3-2, in the wartime and high-growth period the banking sector played an important role in the financial markets and a special relationship between banks and companies developed: the main bank system. Hoshi (1995) argues that the main bank system could only develop because the number of banks had declined in the pre-war period. Large and financially healthy banks are necessary for a system such as the main bank system. The main bank assists and takes responsibility for a member company in financial distress with new lending, reduction in the existing amount of debt, or mediation in activities such as mergers (Caves and Uekusa 1976; Hoshi 1998; Odagiri 1992). The affiliation of a company with its main bank is characterized by a multitude of financial, informational, and managerial relationships. The involvement of the main bank in the operations of the company is illustrated by the Bank loans, Cross-shareholdings, Payment settlement accounts, Information services and supply of management resources, and Bond- issue related services (Aoki 1994).

58 See: Sakamoto (1995), Sheard (1994) and Tanimoto (1994)

96 Section 4-3 Post-war financing and ownership

Bank loans - One characteristic of the main bank system is that companies rely to a high degree on bank loans for their external funding. According to Aoki (1994) this bank borrowing was more than 70% of the total of external finance until 1986. The main bank typically provided the largest part of the company's borrowing. Table 4-5 shows the percentile provision of the loans by financial institutions and the main bank, the latter identified as the city bank that supplies the largest share of loans. Horiuchi (1994) investigates the number of banks that companies use to obtain loans. Large listed companies have an average of 20.8 banks for the total amount of loans. For these companies he finds an average of 7.9 clearing banks, banks that handle daily cash transactions for the companies. If we take these clearing banks as the sum of city banks and regional banks, we see that they provide the companies between 35% and 45% of bank loans. As the main bank's provision of loans amounts to 12.5%-15% during these years, we find that the main bank is responsible for about 33% of these loans. The other 6.9 banks take care of 66% of the loans, and these banks are therefore individually less important to the companies.

Table 4-5 Corporate loans by financial institutions in the period 1977-1991 City banks Regional Year LTCB Trust bank Insurance Others Main bank Others Banks 1977 12.5 16.4 5.7 14.5 16.5 9.8 24.6 1978 12.8 16.7 5.9 14.7 16.3 9.5 24.0 1979 13.3 18.2 6.2 14.5 15.7 8.8 23.2 1980 13.3 18.6 6.3 14.2 15.4 8.7 23.5 1981 13.5 18.9 6.7 14.0 15.0 8.2 23.7 1982 13.6 19.1 7.2 13.9 14.7 7.9 23.5 1983 13.6 19.1 7.8 13.7 14.4 7.5 23.8 1984 14.2 19.6 8.5 13.6 13.8 6.8 23.4 1985 14.8 20.4 9.0 13.4 13.7 6.1 22.6 1986 15.3 21.2 9.1 13.7 13.5 5.7 21.4 1987 15.6 21.8 9.4 14.0 13.3 5.3 20.7 1988 15.7 21.3 9.6 13.7 12.8 5.2 21.7 1989 16.1 21.8 9.4 14.0 12.5 5.3 20.9 1990 15.8 20.2 9.2 13.2 11.7 6.2 23.7 1991 15.0 19.2 9.2 12.3 11.4 8.4 24.4 Source Aoki (1994)

The table also shows that the loan provision of Long-Term Credit banks and Trust banks is at a similar level as the main bank. Aoki (1994) shows that this is primarily caused by the fact that long-term loans at these banks is approximately 20%, while the main bank takes care of long-term loans for a mere average of 7%. In comparison, the Long-Term Credit banks and Trust banks provide only 7% of short-term loans, compared with 21% of the main bank. This confirms that short-term corporate borrowing is the most important characteristic of the main bank relationship. The importance of short-term loans reflects the main bank's oversight role with reference to the ongoing finances, the company's operations and management.

Cross-shareholdings - In the horizontal group, group-members and the main bank are also important shareholders. The main bank is usually one of the top shareholders and most likely to exercise control over the company. According to Kosei Torihiki Iinkai (1992) the

97 Chapter 4 Ownership and corporate governance

group of core companies that collectively own approximately 10-25% possess this control. In principle, the members of the group never sell these shares to any other third party, and will be able to mobilize the shareholdings of the group members in case of an emergency, for example a hostile takeover.

Payment settlement accounts - Another important characteristic of the system is that the company operates a payment settlements account with the main bank. The main bank manages the cash flow receipts and payments of inter-firm transactions, and assumes responsibility for payment of the company's cheques and promissory bills. The bank is also often involved in the foreign exchange business and other banking business of the company. In this way the main bank not only has constant supervision over the financial position of the company, but is basically a part of the company (the "finance-department"). These characteristics affirm the close links between the company and the main bank.

Information services and supply of management resources - The main bank can have a considerable influence on the functioning of the company. It can, for example, provide the client company with information about possible investment or acquisition (assets) opportunities, or introduce potential business partners. Second career main bank associates also often sit as directors or auditors on the client-companies' board of directors. This strengthens the bonding between the main bank and the client-company even more.

Bond-issue related services - In the relationship of the company and the main bank, the latter also plays a significant role in the issues of both domestic and international bonds. The main bank performs the role of the trustee administration, which involves the assessment and custody of collateral and arranging creditors' meetings.

4-3-4 Recent changes in corporate governance and ownership As of the 1970s important changes occurred in the main bank’s corporate governance and the ownership structure of Japanese companies. As indicated in the previous sub-sections, the governance function of the main bank is strongly related to the financing decision of companies. Deregulation created new financing opportunities for companies and bank loans were no longer the cheapest or easiest funding source for companies. The less important role of loans in companies’ financing decision reduced the influence of financial institutions, particularly the main bank (e.g. Schaede 2006). Figure 4-6 summarizes the loan/assets-ratios 59 for manufacturing companies in the period 1960-2006, showing percentages for all companies and those with assets larger than 1 billion yen. The figure indicates that until the mid-1970s the loan/assets-ratio fluctuated between 35% and 40%, but declined as a result of the deregulation process. Especially dependency on loans fell for large companies, indicating that these companies have easier access to other financing methods.

59 Loan consists of short-term loans and long-term loans.

98 Section 4-3 Post-war financing and ownership

0.45

0.40

0.35

0.30

0.25 % 0.20

0.15

0.10

0.05

0.00

0 6 4 0 2 8 0 6 4 6 964 96 970 972 97 978 98 8 986 98 9 994 99 000 002 00 19 1962 1 1 1968 1 1 1 1976 1 1 19 1984 1 1 19 1992 1 1 1998 2 2 2 All companies Large companies

Figure 4-6 Loan/assets-ratio of manufacturing companies in period 1960-2006 Note Loan/assets-ratios for (i) all manufacturing companies, and (ii) manufacturing companies with assets larger than 1 billion yen. Source Ratios calculated based on data retrieved from Ministry of Finance Japan, Policy research institute “http://www.mof.go.jp/1c002.htm”

It is important to realize that, although a lot of Japanese companies have a main bank, this does not apply to each Japanese company. We explained that this relationship evolved from the war-years when the designated banking system was implemented. Companies of strategic importance in the war were assigned to one of the large banks by the government. However, not all Japanese companies were of strategic importance and developed strong links with a bank. Another reason is the vertical keiretsu group. As we explained above, this group is characterized by an important relationship between a subsidiary company and its parent company. The parent company provides the subsidiary company not only with management advice and/or knowledge transfer, but sometimes also with financial resources. This latter function reduces the importance of banks to this type of companies and the possible development of a main bank relationship. Campbell and Hamao (1994) look into the debt/assets-ratio60 of Japanese companies in the period 1970-91 to examine whether affiliation with a main bank had a significant influence on the capital structure of companies. They find that main bank companies have tended to rely more on debt than non-main bank companies in the entire period. The financial deregulation process lowered the debt/assets-ratios for both groups, but the non- main bank companies experienced a stronger decline than companies affiliated with a main bank. A company in a main bank system has a diversity of ties with the main bank and

60 Debt defined as the sum of: (i) short-term borrowings, (ii) current position on long-term debt, (iii) bonds and debentures maturing within one year, (iv) bonds and convertible bonds, and (v) long-term borrowings. Assets are this debt plus the market value of equity

99 Chapter 4 Ownership and corporate governance

other stakeholders, which makes it more difficult to turn to other means for procurement of funds. Although the non-main bank companies may also have had a significant amount of debt, it was easier for these companies to turn to market-based financing instruments and they showed a remarkable change in the use of external funds after the mid-1970s. Schaede (2006) argues that these developments, combined with the change of ownership, have resulted in banks not being able to function as monitors of companies. In chapter 9 we discuss previous research on the monitoring role of the main bank and investigate the main bank’s influence in the merger process with event studies.

The ownership structure of companies started to change in the 1990s. Table 4-6 indicates the percentages of cross-shareholdings and long-term (stable) shareholdings in value and in volume. Cross-shareholdings are between companies that mutually hold each other’s shares, and long-term shareholdings “include ‘shares issued or owned by financial institutions’ and ‘shares held by related listed companies’ (such as parent firms), and measured the long-term shareholding ratio” (Kuroki 2003, p.4).

Table 4-6 Cross-shareholdings and long-term shareholdings in period 1987-2002 Cross-holding Long-term holding Value Unit share Value Unit share 1987 18.4 14.5 45.8 42.5 1988 18.0 14.5 45.7 42.7 1989 17.3 14.6 44.9 42.5 1990 18.0 15.0 45.6 43.1 1991 17.8 14.9 45.6 42.8 1992 17.8 14.8 45.7 42.8 1993 17.5 14.5 45.2 42.0 1994 17.4 14.2 44.9 41.2 1995 17.1 13.9 43.4 39.7 1996 16.3 13.7 42.1 39.0 1997 15.1 13.1 40.5 37.7 1998 13.3 12.4 39.9 36.6 1999 10.6 11.1 37.9 33.9 2000 10.2 10.7 33.0 31.4 2001 8.9 9.1 30.2 30.4 2002 7.4 7.2 27.1 26.0 Source NLI Research Institute

Both types of shareholdings started to decrease in the 1990s, and especially in the latter half. By volume, the cross-shareholding was 15% in 1990, fell to 12.4% in 1998 and to 7.2% in 2002. Long-term shareholdings were at 43.1% in 1990, 36.6% in 1998 and 26% in 2002. The main reason for the decreasing percentages of the two types of shareholdings are that companies with these shareholdings suffered large unrealized losses due to falling share prices after the stock market burst in 1989 (Kuroki 2003). Also, consolidated accounting and reporting requirement of investments at market value put pressure on companies to sell shareholdings in companies and banks. Miyajima and Fumiaki (2005) indicate that selling of bank shares was the result of high risk (non-performing loan problems) and low return; the index for bank shares fell from 100 in March 1995 to 53.8 in February 1999, an index that was considerably lower than the TOPIX at 85.6 as well. The unwinding of shares by companies combined with the banking crisis in 1997 subsequently

100 Section 4-4 Summary

resulted in banks selling their shareholdings in companies (Scher 2001; Miyajima and Fumiaki 2005).

4-4 Summary After Japan opened the country to foreign contact and trade, it realized that its economic development lagged compared to other countries. The government took the initiative in industrial development as it feared to become colonized and realized that individual businessmen did not have sufficient wealth to make the necessary investments. In the 1880s the government sold its companies to businessmen with political relations (seishǀ) and ability to build strong and internationally competitive companies. These businessmen became the leaders of the pre-war zaibatsu groups, conglomerates of various companies, organized in pyramid form, owned by rich families and/or a family-owned holding company. In the post-war period government involvement changed into administrative guidance (gyǀsei shidǀ) of companies. New laws prevented the establishment of holding companies and new corporate structures developed in the post-war period: the vertical keiretsu and horizontal keiretsu. The vertical keiretsu is a group of companies that belong to the same production process of a core company, or parent company. A horizontal keiretsu (kigyǀ shǌdan) is a non-hierarchical group of independent companies in various industrial sectors with origins in the pre-war period. Financial institutions, especially commercial banks, have an important role in these groups and relations between group members are characterized by cross- and stable shareholdings. Corporate governance in the post-war period is characterized by the main bank system, a special relationship of a company with a bank, the so-called main bank. This relationship is characterized by loan provision, cross- shareholdings, payment settlement accounts, information services and supply of management resources, and bond-issue related services. The most important method of financing in the pre-war period was internal finance, but in the 1930s external finance became more important. Regarding external finance it is commonly argued that the use of equity financing decreased in the 1930s, as companies started to rely more on debt finance. We show that this argument is incorrect and that only a gradual decline in equity finance started as of 1942. In the post-war period the stock and bond markets were not efficient and expensive for capital procurement. This led companies to rely to a high degree on bank loans for financing. We look into the developments of share-ownership in the 1950s because previous research argues that individuals started to sell their shares during this period. We show that this conclusion is invalid because it only considers ownership percentages. The absolute number of individuals’ shareholding increased in this period. The lower percentage only indicates that individuals did not, or could not, purchase new share issues. Deregulation of financial markets in the 1980s has resulted in more financing opportunities for companies and a weakening position for the main bank.

101 90

102 5

Mitsubishi groups

In chapter 4 we showed how economic development and informal and formal institutions resulted in the financing decisions of companies with aggregate data. We explained that the holding company played an important role in corporate governance during the pre-war period, and that the main bank was most important in the post-war period. This chapter provides a case study of the Mitsubishi group in order to get more insight in the characteristics of the corporate groups in the pre-war period and post-war period. Section 5-1 discusses the financing decision and ownership structure of the Mitsubishi zaibatsu companies. We examine the post-war financing decision and ownership structure of companies related to the Mitsubishi horizontal keiretsu and the Mitsubishi (main) bank in section 5-2. The Mitsubishi vertical keiretsu is discussed in section 5-3 and the next section summarizes and concludes the chapter. In chapter 7 we will revisit the description of the Mitsubishi zaibatsu and Mitsubishi horizontal keiretsu in order to analyze the transformation according to the institutional model we will explain in chapter 6.

103 Chapter 5 Mitsubishi groups

5-1 Mitsubishi zaibatsu R The holding company decided to change the ownership structure in the period 1930-35, when anti-zaibatsu sentiments were strongest. R Zaibatsu companies increased their capital during the latter 1930s, and the holding company purchased a lot of these shares. R The commonly argued relation between zaibatsu companies’ capital increase and their changing ownership structure is invalid.

The Mitsubishi zaibatsu was founded in 1873 by the entrepreneur Yataro Iwasaki. Iwasaki was born in the Tosa domain, a samurai-controlled region, which is now Kochi prefecture. He started working for the domain in the Nagasaki office of the financial agency and engaged in trading and promotion of products that were produced in the Tosa domain. After the Meiji Revolution61 the Nagasaki office was closed, and in 1869 Iwasaki was transferred to the Osaka office (Osaka Nishinagabori Shokai). The Meiji government banned all domain-related companies, and the Osaka branch was transformed in a private enterprise of which Iwasaki was appointed manager. The next year Iwasaki received a government lease of the Osaka office, which included the trading company, land, buildings and three ships. Following the implementation of the haihan chiken in 1871, which abolished clans and established the prefecture system, Iwasaki took over other enterprises that had been run by the Tosa domain. Iwasaki received land, buildings and ships, and also a silk mill, camphor factory and coal mines that were disposed by the Tosa domain. He established the company Tsukumo Shokai in 1873, later renaming the company Mitsubishi Shokai, and finally, Mitsubishi Mail Steamship Co. Mitsubishi’s direct political involvement started with military transports of the Formosa (Taiwan) Expedition in 1874. The government entrusted thirteen government-owned ships to Mitsubishi in order to facilitate the military transportations for the expedition (these ships had been purchased abroad). In this year the government ordered Mitsubishi to open a steamship line to Shanghai and to develop domestic routes. The government also paid 30 thousand yen and 200 thousand yen as reserve fund and subsidy, respectively. While developing domestic routes, Mitsubishi started other activities such as exchanges, insurance, warehousing and other trade related activities (Shibagaki 1968). The political involvement promoted a monopoly in the management of domestic and international sea routes for Mitsubishi; the marine transport protection policy protected privately owned marine transportation. The government’s ‘First order’ of 1875 transferred the entrusted ships to Mitsubishi free of charge, and the ‘second order’ fixed the period of government assistance funding (250 thousand yen) at fourteen years and transferred another fifteen ships62 to Mitsubishi. The company absorbed another shipping company Kaiso Kaisha that operated between Tokyo and Osaka and by 1875 it had become the largest shipping company in Japan. Having a monopoly in Japan and owning 37 ships, Mitsubishi started routes to Hong Kong (1879) and Vladivostock (1881). Although competition on the Shanghai-route was fierce with P&O Steam Navigation, Mitsubishi drove foreign ship lines out of the market with government support. Mitsubishi's shipping activities in this decade were closely connected with its associations to the Government. In

61 End of the Tokugawa period (1600-1868), start of the Meiji period (1868-1912). 62 The government had bought these ships from a company that had fallen into bankruptcy.

104 Section 5-1 Mitsubishi zaibatsu

the early 1880s, some opinions in the government turned critical of its generosity toward Mitsubishi, and a semi-government shipping enterprise was set up to compete with Mitsubishi. The resulting competition became so fierce that the government had to resolve the situation by merging the two shipping companies into Nippon Yusen Kaisha (NYK) in 1885. Hereupon, Mitsubishi gave up its interests in shipping to turn its business operations to shipbuilding, mining and banking (Fukasaku 1992; Beasley 1963; Allen 1946). During the selling-off of government-operated companies in the 1880s, Mitsubishi was able to obtain the Takashima Coal Mine, the Sado and Ikuno mines, and the Nagasaki Shipbuilding Yard. Nagasaki Shipbuilding Yard was the oldest of the bakufu shipyards established in 1857 by Dutch engineers and, rather than building ships, it facilitated a machinery factory for repair. Mitsubishi Shipbuilding Company was set up in 1917 and it “also diversified its productive activities to build more land machinery and other products especially turbine generators and even steel furniture. In 1934 the firm amalgamated with Mitsubishi Aircraft Company to become Mitsubishi Jukogyo Kabushiki Kaisha (Mitsubishi Heavy Industries Ltd)” (Fukasaku 1992, p.42). In 1893 the holding company, Mitsubishi Goshi, was established. The company implemented an independent accounting system for departments during the period 1908-11, and as of 1911 Mitsubishi Goshi was composed of the following departments: (i) Shipbuilding, (ii) Business, (iii) Real Estate, and (iv) Mines and Coal Mining. The konzern system was adopted in 1917, and turned departments into independent joint stock companies. This process was started with the companies Mitsubishi Shipbuilding & Engineering, Mitsubishi Iron and Steel and Mitsubishi Paper, and the following year (1918) the companies Mitsubishi Mining and Mitsubishi Shoji were made independent. In 1919 Mitsubishi Marine & Fire Insurance and Mitsubishi Bank were established, and in 1920 and 1921 the companies Mitsubishi Internal Combustion Engine Manufacturing Company and Mitsubishi Electric Manufacturing were respectively split off from Mitsubishi Shipbuilding & Engineering (Shibagaki 1968; Okazaki 2001). Due to their important role in the process of militarization leading up to WWII, zaibatsu companies faced an increased need for capital in the 1930s and turned to banks or went public to raise funds. Related to the raising of public funds, Hoshi and Kashap (2001, p.62) argue that the power of the zaibatsu’s headquarters was challenged, and Miyajima (1994) mentions “[W]ith regard to the financing of subsidiary companies, the importance of paid- in capital declined as monetary demand increased during wartime. Holding companies’ stockholding in subsidiary companies decreased as the subsidiary companies’ stock sold on the open market increased” (p.297). Teranishi (1994) explains about the change in corporate financing that internal financing was abandoned “to adapt to the changing industrial structure, and partly to mitigate anti-zaibatsu sentiments” (pp.58-59). “[M]itsubishi was most active with offering shares, it had already started with Mitsubishi Mining and Mitsubishi Bank in respectively 1920 and 1927, and “shares of first-line companies, including holding companies themselves, were offered to the public after 1934” (Teranishi 1994, p.59). Mitsui zaibatsu and Sumitomo zaibatsu started selling shares to the public in respectively 1933-34 and 1934-37 (Teranishi 1994). We find support for the argument of Teranishi regarding the anti-zaibatsu sentiments and the sale of first-line companies’ shares in the 1930s. As explained in chapter 3, in the 1930s various right-wing and anti-zaibatsu movements arose and in 1932 the leader of the Mitsui zaibatsu was assassinated. In this period there were also assassination attempts on other zaibatsu leaders, which undoubtedly forced zaibatsu families to reconsider their

105 Chapter 5 Mitsubishi groups

control over the company. In any event, financing needs were not the only motivation for the families and holding companies. Table 5-1 shows the shareholdings of some Mitsubishi zaibatsu companies and supports our assumption that the anti-zaibatsu sentiment was very important. Mitsubishi Heavy Industries and Mitsubishi Mining are in this respect especially interesting to examine. In 1932 the number of shares of Mitsubishi Heavy was 1 million and its share capital was 50 million yen. The Iwasaki family and the holding company owned 98.6%, which was 49.3 million yen of share capital. In 1935 the percentage of shareholdings of the Iwasaki family and the holding company had decreased to 56.8% of total share capital (30.9 million yen). As described above, it is usually asserted that zaibatsu companies turned to public offering of shares in order to increase funding, herewith reducing the influence of the holding company. This assumption is incorrect for Mitsubishi Heavy as share capital was increased by only 10% in the period 1930-1935. The sale of the holding company, that reduced its ownership by 41.8%-points, had as its main effect that the ownership structure of Mitsubishi Heavy changed. Of course, the reason for the sale by the holding company could be that it was not able to finance the holding of Mitsubishi Heavy’s shares any longer and therefore decided to sell. However, this assumption does not hold if we look at the period 1935-40. In this period Mitsubishi Heavy increased its share capital (and number of shares) more than 4- fold and in 1940 the family and holding company still owned 43.6% of the company. The holding company had increased its shareholdings in Mitsubishi Heavy from 30.9 million yen in 1935 to 104.6 million yen in 1940 to maintain a similar ownership percentage. Table 5-1 shows the same pattern for Mitsubishi Mining; the holding company sells its shares to the public in the period 1920-35, and the holding company purchases shares when share capital is increased in the period 1935-40. These data indicate that the holding company decided to change the ownership of the zaibatsu companies in the period 1930-35, prior to a capital increase. When zaibatsu companies increased their reliance on external funding, the holding company bought shares to maintain an ownership percentage of 40% to 50%. Including ownership by other related companies this resulted in group ownership between 47.2% and 61.4% in 1940. The suggested relation of external finance and change in ownership structure cannot be found for the Mitsubishi zaibatsu companies. The holding company’s desire to prevent losing control over its subsidiaries is also seen by the shareholdings of Mitsubishi Logistics and Mitsubishi Electric at respectively 46.8% and 59.9% in 1940. This was quite a change from 1935 when Mitsubishi Bank was the largest shareholder of Mitsubishi Logistics with 83.5%, and Mitsubishi Shipbuilding had shareholdings of 81.2% in Mitsubishi Electric. The real decline in influence of the holding company occurred in the period 1940-45 when the various war-related regulations were implemented. Ownership of Mitsubishi Heavy by the holding company and related companies, for example, had fallen to 25.1% in 1945.

106 Section 5-1 Mitsubishi zaibatsu 61.4 53.1 58.8 50.3 57.8 47.8 76.0 47.5 78.5 88.1 60.2 44.6 59.9 47.3 47.2 Total 0.8 43.9 1.0 97.0 0.4 32.7 0.20.6 25.3 99.2 2.00.7 25.1 99.2 2.9 44.3 1.0 94.2 0.9 98.2 14.7 63.7 25.2 48.1 Other Related 0.8 Glass Asahi 0.6 0.5 Trading ing 81.2 6.7 Shipbuild 0.3 1.0 0.5 1.3 1.3 Heavy Mining 5.1 9.8 0.3 25.7 Bank Trust 0.3 83.5 Bank Fire Tokyo Meiji Life & Iwasaki Holding 55 56.1 2.7 15 96.710 7.7 30 23.4 8.7 14.3 6.7 41 16.1 8.2 13.4 40 22.1 9.4 18.8 3020 59.9 78.5 30 0.2 20 46.8 10.0 3.1 50 41.2 1.3 1.3 (million yen) Share Capital 1940 100 43.21940 4.6 240 8.5 1940 43.6 200 4.2 43.5 2.5 3.6 0.7 1935 19351935 1001935 56.8 0.4 1940 1943 1935 1940 1925 1935 1940 1940 1929* 100 49.0 1928*1932* 30 22.9 501920* 98.6 50 98.5 1927* 50 96.0 1945**1945** 135 32.3 301945** 25.1 1,0001945** 23.1 1945** 407 43.1 100 41.4 1945** 120 44.6 1945** 20 47.3 Without * indicates that data* is Indicates information retrieved on from, (i) amount andGaisha of calculated Nenkan, share and based capital (ii) is on, ownership retrieved the percentages from are** respective the from Indicates company issue Teranishi that history (1994), we of as Table retreived noted Kabushiki 2.4 data indicated on Gaisha from in page Asajima Nenkan. the 60). (1986, Kabushiki Table 14, p.224). Table 5-1 Ownership of Mitsubishi zaibatsu companies Notes and sources: ï ï ï Mitsubishi Bank Mitsubishi Trust Mitsubishi Heavy Mitsubishi Mining Mitsubishi Trading Mitsubishi Logistics Mitsubishi Electric Asahi Glass

107 Chapter 5 Mitsubishi groups

Another interesting aspect of the pre-war ownership structure is the importance of Meiji Life Insurance and Tokyo Fire and Marine as related shareholders. As of 1940 they have shareholdings in Mitsubishi Bank (13.1%), Mitsubishi Trust Bank (23%), Mitsubishi Heavy (6.7%), Mitsubishi Mining (4.3%), Mitsubishi Trading (2.6%), Mitsubishi Logistics (13.1%), and Asahi Glass (in 1935 at 28.2%). Tokyo Fire and Marine was also the most important investor in group companies of the Mitsubishi zaibatsu pyramid with ownership in 23 companies in 1930. Mitsubishi Bank was second with shareholdings in ten, followed by Mitsubishi Mining and Mitsubishi Trading with respectively shareholdings in eitgh and seven group companies. For the large companies in table 5-1 the shareholdings by Mitsubishi bank, Mitsubishi Trust bank and related companies were less important. The only exception was Mitsubishi Trading having Mitsubishi Heavy, Mitsubishi Mining and Asahi Glass as shareholders in 1940.

5-2 Mitsubishi horizontal keiretsu R Resulting from the main bank system, companies have a strong relationship with either the Mitsubishi Bank or the Mitsubishi Trust Bank. R Corporate cross-shareholdings in the horizontal keiretsu are very limited and concentrated in the six most important companies. We show their shareholding relationships with our “Mitsubishi Keiretsu Diamond”. R The diamond companies and financial institutions have shareholdings in each other and important subsidiaries.

As chapter 4 explained, the post-war Japanese financial system is characterized by the horizontal keiretsu and the main bank system. Until recently, these two concepts were accepted as important characteristics, but Miwa and Ramseyer (2001) have questioned their importance and even their existence. They examine the relation between the main bank and group companies, and the cross-shareholdings between group companies. They do not find strong evidence to support the importance of either the main bank or horizontal keiretsu and conclude that both are a myth. Miwa and Ramseyer (2001) claim that the Japanese keiretsu is a “figment of academic imagination”, by arguing that the concepts used for cross-shareholdings and main bank are trivial. They explain that ideas of the largest amount of loans, the monthly luncheon of presidents, and the companies in which presidents held equity positions, are not a strong enough base to qualify companies as being part of a main bank system or not. Their assertion provides an interesting starting point to examine the horizontal keiretsu because it is often argued to prevent hostile takeovers. We replicate the study of Miwa and Ramseyer (2001) on a slightly smaller scale and investigate shareholdings and loan relations of 55 companies in the Mitsubishi keiretsu for the year 1975, and of 24/25 companies in the years 1988 and 1998. Our data confirm their findings with reference to the low number cross-shareholdings, but we find strong evidence for the existence of the horizontal keiretsu and main bank system.

5-2-1 Data and Methodology Our starting sample is the selection of Mitsubishi keiretsu companies as identified by Nakatani (1984, p.233). Nakatani investigates profitability of keiretsu groups and because we are interested in the shareholdings and loans between the group companies, we add

108 Section 5-2 Mitsubishi horizontal keiretsu

Mitsubishi Bank, Mitsubishi Trust Bank, Meiji Life Insurance, Tokyo Fire & Marine, Mitsubishi Trading, Mitsubishi Estate, and Mitsubishi Heavy Industries. Our final sample consists of 55 companies and four financial institutions. We obtain shareholder and loan data of the sample companies for the year 1975 from the publication Kigyǀ Keiretsu Soran, which shows the largest 20 shareholders of the sample companies and the total outstanding loans at the various financial institutions.

Table 5-2 Loan provision and shareholdings by Mitsubishi keiretsu financial institutions Loans Shareholdings No. Comp. Mean No. Comp. Mean Bank 52 16.5 51 4.8 Trust Bank 52 13.4 48 3.0 Life Insurance 17 1.8 42 3.5 Cas. Insurance 2 0.0 44 2.3 Total 31.7 11.3 Notes Our sample consists of 55 companies; the Mitsubishi keiretsu companies as identified by Nakatani (1984, p.233), and Mitsubishi Bank, Mitsubishi Trust Bank, Meiji Life Insurance, Tokyo Fire & Marine, Mitsubishi Trading, Mitsubishi Estate, and Mitsubishi Heavy Industries. We obtain shareholder and loan data of the sample companies for the year 1975 from the publication Kigyǀ Keiretsu Soran, Sources Nakatani (1984, p.233) and Kigyǀ Keiretsu Soran, for the year 1975.

We first look into the provision of loans by the group’s financial institutions. Mitsubishi Bank and Mitsubishi Trust Bank provide 51 companies with loans, and Meiji Life Insurance has loans outstanding at 17 companies. We find one company with loans from Mitsubishi Bank, one company with loans from Mitsubishi Trust Bank only, and two companies do not have loans outstanding at any of the group’s financial institutions. Average loan percentage for Mitsubishi Bank is at 16.5%, for Mitsubishi Trust bank at 13.4% and for Meiji Life Insurance at 1.8%. The companies also rely on other financial institutions for loans such as Taiyo Kobe Bank (38 cases with an average of 1.9%), Tokai Bank (37 cases with an average of 2.7%), Industrial Bank of Japan (35 cases with an average of 4.2%) and the Long-term Credit Bank (34 cases with an average of 5.0%). All sample companies with loans had a group’s financial institution as their most important source for funding. One of the most important aspects of the main bank system, as part of the horizontal keiretsu, is that the financial institutions have shareholdings in the companies it provides loans to. Mitsubishi Bank has shareholdings in 51 companies, Mitsubishi Trust in 48 companies, Meiji Life Insurance and Tokyo Kasai in respectively 42 and 44 companies. In 34 companies all 4 financial institutions have share-ownership (average of 16.6%), and in each company either Mitsubishi Bank, Mitsubishi Trust Bank or both have shareholdings. Of the companies in which not all financial companies have ownership, in 19 companies there are two or three financial institutions with shareholdings, and there is only one shareholder in two companies. Miwa and Ramseyer (2001) investigate whether the financial institutions act cohesively in loan-provision to test whether keiretsu groups exist. We replicate this test and as table 5- 3 shows, loan provision by the Mitsubishi Bank is negatively correlated with loan provision by the Mitsubishi Trust Bank and Meiji Life Insurance, similar to Miwa and Ramseyer (2001).

109 Chapter 5 Mitsubishi groups

Table 5-3 Loan correlation Mitsubishi keiretsu financial institutions Bank Trust Bank Life Ins. Bank 1.000 Trust Bank -0.266 a 1.000 Life Ins. -0.120 0.204 1.000 Notes “a” indicates significance at 0.10 level. Our sample consists of 55 companies; the Mitsubishi keiretsu companies as identified by Nakatani (1984, p.233), and Mitsubishi Bank, Mitsubishi Trust Bank, Meiji Life Insurance, Tokyo Fire & Marine, Mitsubishi Trading, Mitsubishi Estate, and Mitsubishi Heavy Industries. We obtain shareholder and loan data of the sample companies for the year 1975 from the publication Kigyǀ Keiretsu Soran. Sources Nakatani (1984) and Kigyǀ Keiretsu Soran for the year 1975.

Unlike Miwa and Ramseyer (2001), we interpret the results in table 5-3 not as evidence that the group’s financial institutions did not act cohesively. In contrast, we argue that it needs to be considered as cohesive action; if a company within the group needs loans, it seems inefficient and unnecessary that the loans are retrieved equally from each financial institution. The group element should be seen in the fact that in some cases primarily financial institution A provides loans, and in other cases primarily financial institution B. This can be examined by analyzing the loans provided to the 55 companies in our sample by Mitsubishi Bank and Mitsubishi Trust Bank.

90%

80%

70% s

60%

50%

40%

30% Loans as percentage of total loan 20%

10%

0% 135791113151719212325272931333537394143454749515355 Mitsubishi Bank Mitsubishi Trust Bank

Figure 5-1 Loans from Mitsubishi Bank and Mitsubishi Trust Bank in 1975 Notes Our sample consists of 55 companies; the Mitsubishi keiretsu companies as identified by Nakatani (1984, p.233), Mitsubishi Trading, Mitsubishi Estate, and Mitsubishi Heavy Industries. The banks are: Mitsubishi Bank and Mitsubishi Trust Bank. We obtain loan data of the sample companies for the year 1975 from the publication Kigyǀ Keiretsu Soran, Sources Nakatani (1984) and Kigyǀ Keiretsu Soran for the year 1975.

Figure 5-1 shows the loan percentages of the total outstanding loans for each company, starting with the company that had the highest loan provision by Mitsubishi bank. We find that some companies have similar loan percentages at both financial institutions, but that a majority of the companies either relied on Mitsubishi Bank or Mitsubishi Trust Bank for

110 Section 5-2 Mitsubishi horizontal keiretsu

the majority of its loans. The financial companies within the group thus acted cohesively by assigning companies to particular financial institutions within the group, reducing the necessity to provide loans at the same level to all companies. Next, we investigate the shareholdings between the keiretsu companies. Excluding financial institutions, we find 94 shareholdings in other group companies and these holdings are primarily held by 16 companies. The number of cross-shareholdings within the group is low with only twelve cases, corresponding with the findings of Miwa and Ramseyer (2001). The average ownership by other group companies is 11.07% and by financial institutions 13.58%, bringing the average group shareholding to 24.65%. Of these companies, ownership by group companies and financial institutions exceeds 50% in seven companies and 33.3% in 16 companies. Shareholdings of independent and other financial institutions increase the average shareholding to 43.52%. Terming these financial institutions as “stable shareholders” increases the number of companies with shareholding over 50% and 33.3% to respectively 21 and 43 companies.

Table 5-4 Group ownership in Mitsubishi keiretsu Group Group Sub- Indep. Sub- Other Total companies finance total finance total finance Average shareholding(%) 11.07 13.58 3.12 15.75 24.65 27.77 43.52 Number of companies Shareholding > 50% 7 8 21 Shareholding > 33.3% 16 20 43 Notes Our sample consists of 55 companies; the Mitsubishi keiretsu companies as identified by Nakatani (1984, p.233), and Mitsubishi Bank, Mitsubishi Trust Bank, Meiji Life Insurance, Tokyo Fire & Marine, Mitsubishi Trading, Mitsubishi Estate, and Mitsubishi Heavy Industries. We obtain shareholder and loan data of the sample companies for the year 1975 from the publication Kigyǀ Keiretsu Soran, Sources Nakatani (1984) and Kigyǀ Keiretsu Soran for the year 1975.

Of the 16 companies with shareholdings in other group companies, we investigate the companies with shareholdings in seven or more companies. We get a sample of six companies in which the cross-shareholdings are also concentrated with ten cases. The companies in this group are Mitsubishi Trading (29 shareholdings), Mitsubishi Estate (7), Mitsubishi Heavy (17), Asahi Glass (10), Mitsubishi Electric (10) and Mitsubishi Petrochemical (10). When we link the cross-shareholding relationships between these companies, we discover the Mitsubishi Keiretsu Diamond63 asshowninFigure5-2.

63 Our analysis of the shareholding relationships between companies in the Mitsubishi keiretsu resulted in figure “Mitsubishi keiretsu diamond” and shows the cross-shareholdings of Mitsubishi companies. As the Mitsubishi symbol is a diamond, we have termed the figure as such and it indicates the relationships between the most important companies in the keiretsu.

111 Chapter 5 Mitsubishi groups

Heavy Petrochemical Asahi Glass

Trading Estate

Electric

Figure 5-2 Mitsubishi Keiretsu Diamond Notes Our sample consists of 55 companies; the Mitsubishi keiretsu companies as identified by Nakatani (1984, p.233), and Mitsubishi Bank, Mitsubishi Trust Bank, Meiji Life Insurance, Tokyo Fire & Marine, Mitsubishi Trading, Mitsubishi Estate, and Mitsubishi Heavy Industries. We obtain shareholder and loan data of the sample companies for the year 1975 from the publication Kigyǀ Keiretsu Soran, Our sample consists of 16 companies that have shareholdings in other group companies. The figure shows the 6 companies in which the cross-shareholdings are concentrated with 10 cases.

The diamond indicates the cross-shareholdings of, and between, Mitsubishi Trading, Mitsubishi Estate, Mitsubishi Heavy, Asahi Glass and Mitsubishi Electric. The shareholdings in Mitsubishi Trading by the other diamond companies are highest with 10.2% and lowest in Mitsubishi Petrochemical with 1.7%. Shareholdings in the other companies range between 5.0% and 5.8%, of which Mitsubishi Trading has the most extensive shareholdings with 10.8%. These six companies are also important shareholders of the group’s financial institutions and together they own 13.5% and 11.0% in respectively Mitsubishi Trust Bank and Mitsubishi Bank. The financial companies own shares in all diamond companies, ranging from a total of 26% in Asahi Glass to 11.1% in Mitsubishi Electric. Although Mitsubishi Trading is the only diamond company with ownership in Asahi Glass, including the financial institutions brings the total to 31.7%. The financial institutions also have important shareholdings in Mitsubishi Trading (24.5%), Mitsubishi Petrochemical (19.9%), and each other. The financial institutions are clearly at the apex of the keiretsu structure and with reference to shareholdings in group companies, Mitsubishi Bank and Tokyo Casualty are most important.

100

112 Section 5-2 Mitsubishi horizontal keiretsu

Table 5-5 Ownership of/by diamond companies and financial institutions Comp. Tra- Asahi Petro- Sub- Trust Non- Sub- Heavy Estate Electr. Bank Total Shareh. ding Glass chem. total Bank Life total MB Trading 2.4 5.6 1.1 1.7 10.8 2.0 3.7 2.3 8.0 18.8 MB Heavy 5.6 1.5 2.6 4.1 3.7 3.2 2.1 9.1 13.1 Asahi Glass 1.5 1.0 1.7 1.0 3.6 2.0 2.5 1.8 6.3 9.9 MB Estate 1.1 1.0 0.8 0.7 2.5 0.9 1.5 2.4 4.9 MB Electric 2.0 1.4 0.8 2.2 1.2 1.7 2.9 5.1 MB Petrochem. 0.0 1.2 0.8 2.0 2.0 Sub-total 10.2 5.8 5.6 5.1 5.0 1.7 11.0 13.5 6.1 MB Bank 7.9 5.8 8.3 4.0 3.4 5.5 26.8 5.8 5.8 32.7 MB Trust 4.1 2.6 5.2 2.7 1.1 2.9 14.5 1.5 2.6 3.1 7.2 21.7 Tokyo Casualty 4.9 3.7 9.1 4.0 5.0 7.8 29.6 5.9 7.7 4.8 18.4 48.0 Meiji Life 7.7 3.2 3.4 4.0 1.6 3.7 15.8 4.7 1.7 6.4 22.2 Sub-total 24.5 15.3 26.0 14.6 11.1 19.9 12.0 12.0 13.7 Total 34.7 21.0 31.7 19.7 16.1 21.6 23.0 25.5 19.9 Notes Our sample consists of 16 companies that have shareholdings in other group companies. The “Mitsubishi Keiretsu Diamond” shows the 6 companies in which the cross-shareholdings are concentrated with 10 cases. We obtain shareholder and loan data of the sample companies for the year 1975 from the publication Kigyǀ Keiretsu Soran,

Regarding shareholdings, other than the cross-shareholdings in the keiretsu diamond, we find large differences in ownership percentages with a minimum of 0.3% and maximum of 61.3%. Mitsubishi Trading is clearly at the apex of the keiretsu with shareholdings in 24 other group companies, being the only shareholder of the diamond companies in eleven companies at an average ownership of 1.5%. This low percentage should not be interpreted as evidence for these companies’ unimportance to the group though. The average group shareholding increases to 29% when other group companies and financial institutions are included. The importance of the shareholdings by the diamond companies is evident when comparing the companies in which they have shareholdings and those in which they do not. In 37 companies one or more of the six companies have shareholdings at an average of 10% and, including other group companies and the financial institutions, group ownership in these companies increases to 31.2%. In the 18 companies in which the six do not have shareholdings the average of the group is only 10.23% with financial institutions being very important with shareholdings of 10.16%. As we explained in the previous chapter, it is argued that the cross-shareholding started to disappear with the deregulation of Japan’s financial markets. Table 5-6 shows that group ownership of the diamond companies and the three financial companies is indeed decreasing. Average group ownership in companies fell from 24.1% in 1975, to 22.3% in 1988, and subsequently to 20.4% in 1998. The decline in 1988 was caused by falling ownership of companies, and the lower ownership in 1998 by decreasing shareholdings of financial institutions. Average shareholdings in financial institutions amounted to 22.8% in 1975, 21.5% in 1988 and 18.2% in 1998. Over the period we see a continuing fall in ownership of companies, by 0.7% in 1975-88 and 1.6% in 1988-98. Ownership of financial companies only shows a strong decline by 1.8% in the period 1988-98.

101

113 Chapter 5 Mitsubishi groups

Table 5-6 Cross-shareholdings of diamond companies and financial institutions 1975 1988 1998 Trading 34.7 28.7 27.4 Heavy 21.0 17.8 14.4 Asahi Glass 31.7 26.0 24.3 Estate 19.7 23.5 20.2 Electric 16.1 16.1 14.4 Chem. 21.6 21.7 21.9 Av. Comp. 24.1 22.3 20.4 Bank 23.0 21.2 15.3 Trust Bank 25.5 22.0 20.6 Non-Life 19.9 21.4 18.6 Av. Fin. 22.8 21.5 18.2 Sources Own data-set for 1975. Publication Nippon Kigyǀ Shǌdan for the years 1988 and 1998.

5-3 Mitsubishi vertical keiretsu R Companies in the horizontal keiretsu are in most cases the core company in a vertical keiretsu. R Some companies in the horizontal keiretsu are part of the vertical keiretsu of another horizontal keiretsu company. R Previous research on horizontal keiretsu can be biased because of the influence of the vertical keiretsu relation between horizontal keiretsu companies.

In this section we look into the vertical keiretsu groups of companies within the Mitsubishi horizontal keiretsu. As we already explained in chapter 3, vertical keiretsu groups are an important characteristic in Japan and exist in each industrial sector. Most of the large companies of the Mitsubishi horizontal keiretsu are therefore also the parent company of a vertical keiretsu. In this section we examine the number of companies in a vertical keiretsu and investigate whether the companies identified as being part of a horizontal keiretsu, are also part of each others’ vertical keiretsu relationship. We use a publication that indicates the corporate groups in the industrial sectors and shows the vertical keiretsu of the largest company in each sector. As the publication only gives information on the largest companies, detailed information is not given for each Mitsubishi company. Table 5-7 shows the number of subsidiaries and affiliated companies for six Mitsubishi companies. Large vertical keiretsu groups are those of Mitsubishi Shoji with 574 subsidiaries and 206 affiliated companies and Nippon Shipping with 329 subsidiaries and 32 affiliated companies. The high number of companies in the vertical keiretsu reinforces the suggestion by Westney (2001) that a Japanese company cannot be as a separate entity.

102

114 Section 5-3 Mitsubishi vertical keiretsu

Table 5-7 Shareholdings in Mitsubishi vertical keiretsu in the year 1988 Shareholdings Shareholdings exlcuding vertical No. of No. of Belongs to vertical Company keiretsu Subs. Aff. keiretsu of Corp. Total Corp. Total MB Construction 62.3 70.2 MB Materials 25.8 33.7 MB Plastics 48.0 55.1 MB Chemical 0.5 7.6 MB Automotive 38.3 50.4 MB Heavy Ind. 14.4 26.6 MB Cable Industries 32.9 46.6 MB Materials 3.7 17.4 MB Shindoh 32.7 42.0 MB Materials 5.1 14.4 MB Storage 18.3 40.8 18.3 40.8 MB Steel Mfg. 17.0 34.3 17.0 34.3 MB Kakoki 15.0 34.9 15.0 34.9 MB Shouji 10.4 31.3 574 206 10.4 31.3 MB Paper Mills 9.0 28.7 9.0 28.7 Nikon 7.4 26.0 7.4 26.0 MB Materials Corp. 7.1 20.2 114 24 7.1 20.2 Asahi Glass Company 6.7 26.6 200 46 6.7 26.6 Nippon Shipping 6.5 22.2 329 32 6.5 22.2 MB Gas Chemical Comp. 6.0 21.8 6.0 21.8 MB Real Estate 5.9 21.9 5.9 21.9 MB Rayon 5.8 19.2 5.8 19.2 Kirin Brewery 5.4 21.5 5.4 21.5 MB Chemical Corp. 5.2 23.4 168 113 5.2 23.4 MB Heavy Industries 4.4 15.4 128 53 4.4 15.4 MB Electric Corp. 3.6 15.0 3.6 15.0 Average 15.3 32.7 8.8 22.8 Sources Publication Nippon Kigyǀ Shǌdan for ownership percentages; vertical keiretsu information from Kigyǀ gurupu to gyoukai chizu.

We examine whether a company in the horizontal keiretsu is part of another company’s vertical keiretsu and first look into the ownership of the companies in the horizontal keiretsu. The first and second columns of the table indicate the shareholdings of group companies and all group companies (including ownership of the group’s financial institutions) respectively. The average for company ownership is 15.3% and for total ownership the average is 32.7%. The table indicates corporate ownership percentages exceeding 30% for the following five companies: (i) Mitsubishi Construction at 62.3%, (ii) Mitsubishi Plastics at 48%, (iii) Mitsubishi Shindoh at 32.7%, (iv) Mitsubishi Cable Industries at 32.9% and (v) Mitsubishi Motors at 38.3%. All these companies belong to the vertical keiretsu of another company in the horizontal keiretsu. The companies Mitsubishi Construction, Mitsubishi Shindoh and Mitsubishi Cable Industries belong to the vertical keiretsu of Mitsubishi Materials, Plastics belongs to Mitsubishi Chemical, and Mitsubishi Motors belongs to the vertical keiretsu of Mitsubishi Heavy Industries. In the two columns on the right of table 5-7 we show the ownership percentages excluding the shareholdings of parent companies in the vertical keiretsu. The average shareholding of group companies falls from 15.3% to 8.8%, and the average total shareholdings from 32.7% to 22.8%. None of the companies have corporate ownership exceeding 30%. These findings are interesting with reference to previous research on the influence of the horizontal keiretsu. The results of previous research can be biased because they included relations between a parent company and a subsidiary in a vertical keiretsu

103

115 Chapter 5 Mitsubishi Groups

instead of separate entities that had cross-shareholdings.

5-4 Conclusions This chapter examined the financing decision and ownership structure of companies in the pre-war Mitsubishi zaibatsu and post-war Mitsubishi keiretsu. Prior to the war the holding company was most important and owned in most cases more than 90% of the shares in the main group companies. During the 1930s the zaibatsu families started to change the ownership structure because of the anti-zaibatsu pressures. In this period the first signs of group shareholdings appeared, especially the insurance companies were important and in some cases other related companies. Mitsubishi Bank and Mitsubishi Trust Bank were relatively unimportant shareholders and also during the war period 1937-45, when loan financing increased, banks did not develop into important shareholders. The main bank system and the horizontal keiretsu developed after the war. In these systems financial institutions are at the apex with loan provision and share ownership of all group companies. Similar to the pre-war zaibatsu, insurance companies are important shareholders in group companies in the post-war system. A new development can be seen in Mitsubishi Bank and Mitsubishi Trust Bank becoming important shareholders of group companies. We did not find a lot of cross-shareholdings in our sample and they are concentrated in six companies, which we termed as the ‘Mitsubishi keiretsu diamond’ companies. The financial institutions have close relations with the keiretsu diamond companies through cross-shareholdings and the two groups together have extensive ownership in other (important) group companies. The diamond companies and the four financial institutions were all direct subsidiaries of the pre-war Mitsubishi holding company and Iwasaki Family. This indicates that the pre-war apex companies remain important as apex companies in the post-war period. We found that companies in the horizontal keiretsu are in most cases the core company in a vertical keiretsu, but some companies in the horizontal keiretsu are part of the vertical keiretsu of another horizontal keiretsu company. Previous research on the horizontal keiretsu might be influenced by relations between a parent company and a subsidiary, and results could therefore be biased.

In chapter 7 we will analyze the findings of the Mitsubishi groups in the pre-war and post- war periods in relation to the informal and formal institutions and economic development, as described in chapters 3 and 4. In our institutional model we pay special intention to the historical and context-specific background. We examine the development of the vertical keiretsu and the transformation of ownership structure from the pre-war zaibatsu to the post-war horizontal keiretsu. In the next chapter we will first introduce the institutional model that will be used for this analysis.

104

116 6

Institutional model for M&A

This chapter presents the institutional model for M&A activity. The background information, as discussed in chapters 3 through 5, is used to design the model. We integrate Japan’s informal institutions and formal institutions, and their subsequent influence on companies’ financing decision, ownership structure, and formation of corporate groups. Our model offers a framework to examine the institutional elements with a dynamic perspective, and their influence on M&A activity. The model is inspired by the Williamson 4-level model (Williamson 1998), Williamson 3-level model (Williamson 1996), insights from North (1990, 2005) and Greif (2006). Williamson provides us with a good theoretical framework to investigate the vertical keiretsu in Japan, and North indicates the importance of path dependence in institutional change. Greif presents strong evidence to investigate the significance of groups in a society, and to include historical and context-specific background of institutional elements in the analysis. In section 6-1 we start with an explanation of the framework of the institutional model, and in the following three sections we discuss the four levels. The levels are respectively the informal institutions, formal institutions, governance structure and M&A activity, and individual actors. Section 6-5 explains institutional change and the next section concludes the chapter.

105

117 Chapter 6 Institutional model for M&A

6-1 Framework of institutional model R The institutional model for M&A consists of five boxes at four levels: informal institutions at level one, formal institutions at level two, governance structure and M&A activity at level three, and individual actors at level four. R In our institutional model we open the neoclassical black boxes of governance structure and individual actors. R Motivation is important in the model because it induces an individual to follow social rules (or not) and results in the individual’s behaviour. R Being a member of a group results in a shared mental model that has an important influence on the motivation of an individual, and therefore functions as an institutional element. R The elements in the model need to be studied against a historical and context-specific background, as this is the information that actors of a society share and internalize.

In this section we explain our institutional model for M&A activity. Before we look into the model we explain institutions. North (1994) defines institutions as: “[T]he humanly devised constraints that structure human interaction. They are made up of formal constraints (e.g., rules, laws, constitutions), informal constraints (e.g., norms of behavior, conventions, self-imposed codes of conduct), and their enforcement characteristics” (p.360). Greif (2006) mentions that institutions are the equilibrium phenomena of the interaction of social rules and motivation, and consist of institutionalized rules. As such, “[I]nstitutions are shaped by a society’s social and cultural heritage, and they contain norms and internalized and behavioural beliefs” (Greif 2006, p.380). Institutionalized rules evolve from the social rules when individuals are motivated to follow them, and they are commonly believed to be followed by others. A social rule can only become institutionalized and part of an institution when it is self-enforcing. “[B]ehavior is self- enforcing in that each individual finds it best to follow the institutionalized behavior that, in turn, reproduces the institution in the sense that the implied behavior confirms the associated beliefs and regenerates the associated norms” (Greif 2006, p.16). This implies that each individual decides, based on his private knowledge and information, to follow the expected behaviour whilst expecting others to behave in a similar way. Institutional elements are formed by the institutionalized rules and provide the micro-foundations of individuals’ behaviour. As such, institutional elements consist of the shared cognitive system, coordination and information, and “inherited from the past are the properties of societies and individuals, history - encapsulated in institutional elements - influences selection among alternative institutions in new situations” (Greif 2006, p.380). Figure 6-1 shows our institutional model for M&A activity and indicates that informal and formal institutions are located at levels one and two respectively. Informal institutions constrain the three other levels and consist of culture, norms, and the ie system (as described in section 3-1). Formal institutions constrain governance structure, M&A activity and individual actors. The formal institutions are formed by the company law, securities exchange law, Antimonopoly Act, and government policies. These laws have an important influence on M&A activity as they regulate the shareholder rights, tender offer procedures, and companies’ organizational structure respectively (as described in section 3-2). In our model we treat informal and formal institutions as exogenous parameters.

106

118 Section 6-1 Framework of institutional model

Informal institutions are held constant and we consider a change in formal institutions as an exogenous parametric change that brings about new transactions and necessitates a search for new institutions. With the model we focus on the process of change and argue that new institutions not only reflect environmental factors, but also the impact of institutional elements inherited from the past. The parameters form an important part of institutions and influence the endogenous variables; governance structure, M&A activity, and individual actors.

Informal institutions Religion and tradition; - Shinto, Buddhism, Confucianism Ⴓ L1 Norms/values (Constitution of Seventeen Articles): Ⴓ - Harmony & Tolerance, Modesty, Group, Trust Ie system and dǀzoku system

Formal institutions Company Law Securities Exchange Law L2 Antimonopoly Act Ⴓ Government policies (competition, economic, financial)

Governance structure

Horizontal Vertical Trade L3 M&A keiretsu keiretsu association activity

Company

Individual actors

Motivation Behavioral beliefs Internalized beliefs Internalized norms

L4 Individual Shared mental mental maps maps

Shared cognition Mental maps - Habits, routines - Cultural learning - Experience

Figure 6-1 Institutional model for M&A

107

119 Chapter 6 Institutional model for M&A

M&A activity, consisting of greenmail, hostile tender offers, and mergers, is positioned at level three. Governance structure is also placed at level three, and individual actors at level four. M&A activity, governance structure, and individual actors are linked to each other with bi-directional solid arrows, because they have a mutual influence on each other, as will be explained in section 6-3. The neoclassical box ‘Governance structure’ is opened because, in order to understand M&A activity, we argue it is important to analyze: (i) the company, and (ii) corporate groups such as horizontal keiretsu, vertical keiretsu, and trade associations. The box ‘company’ represents a company’s ownership and governance structure, rather than the neoclassical interpretation of a production function that determines the price and output level of a product in order to maximize profits. Informal and formal institutions have an important impact on the ownership and governance structures of companies, and the motivation to form corporate groups. We take a historical and context-specific perspective to examine the dynamic development of companies and corporate groups. Our model shows that level three is an outcome of the dynamic interactions between economic development, (foreign) competition, and changes in informal and formal institutions that govern motivations and behaviour of individuals. The box ‘Individual actors’ consists of three layers: (i) shared cognition, (ii) individual mental maps and shared mental maps, and (iii) motivation. We open the black box of individual actors because we think the traditional neoclassical theory does not sufficiently explain the behaviour of actors. The neoclassical theory revolves around the allocation of scarce resources among alternative ends, emphasizing equilibria. These equilibria are the solutions of maximization problems of self-interested individuals who make fully rational decisions and have rational preferences among outcomes that can be identified and associated with a value. Our model indicates that behaviour of actors is the result of the interaction between the mentioned three layers. The layers should be analyzed separately in order to understand why particular individual actors initiate M&A activity. Informal and formal institutions provide individuals with shared cognition, which is exogenous to each individual whose behaviour it influences. The informal and formal institutions also result in an individual actor’s mental map and motivation. The mental map enables, guides, and motivates the individual actor to follow a specific behaviour. Depending on his motivation, each individual actor will behave in a particular manner in response to the institutional elements implied by others’ behaviour and/or expected behaviour. Belonging to a group or not, will also have an important influence on the individual actor’s behaviour (shared versus individual mental map). The behaviour of actors, constrained by informal institutions and formal institutions, influences governance structure and M&A activity. Changes that affect the economic activities in a society, such as increased foreign competition, may also influence the actions of actors at the micro-level through the strengthening or weakening of social norms, social relationships, and cultural attributes. There is feedback between the levels of the model. For example, new formal institutions will influence the informal institutions, and companies, corporate groups, and individual actors will try to influence formal institutions. Although these strategic actions can be very important and are interesting to analyze, their analysis is not feasible in the scope of the current study. The model therefore does not have feedback from governance structure and individual actors to formal institutions. As described above, we have only added solid arrows between governance structure, M&A activity, and individual actors.

108

120 Section 6-2 Informal and formal institutions

In chapter 7 we will analyze hostile takeovers in post-war Japan according to our model. We think that this model is better suited for this analysis than methods used in previous research; previous research provides explanations with a static perspective and treats the institutional elements as exogenous variables. The two findings on hostile takeovers in previous research are (i) the influence of a company’s stable and cross-shareholdings, and (ii) the specific Japanese culture. Regarding the stable and cross-shareholdings, the influence of shareholders (individual actors) on hostile takeovers is looked into by analyzing it at a specific point in time. The analysis does not look into the historical and context-specific background, which necessitates including the motivation of individual actors to be(come) shareholder, and examining how/why (changes in) formal institutions influence the stable and cross-shareholdings. Similar to the shareholdings, the argument of the specific Japanese culture does not explicitly identify the motivation of actors to act according to the ‘harmony’ characteristic; there is no distinction made between individual actors. Both arguments are therefore isolated and self-terminating, rather than exhaustive, once a critical list of variables is determined. The focus lies on the reason why hostile takeovers do not occur, in stead of identification of the nature of hostile takeovers and the motivation of the actors to initiate them. The two arguments therefore do not differentiate between the two forms of hostile takeovers, hostile tender offers and greenmail, despite the large difference in their nature and the actor’s motivation. In our model we solve these problems by taking the dynamic development of various elements within the Japanese society into consideration. We do not pre-define dependent and independent variables, but focus on the full complexity of governance structures and individual actors. The informal and formal institutions need to be considered in relation to the motivation of individuals. Another important feature of our model is that it looks into the historical and context-specific background of the informal and formal institutions, groups, and the economy. These factors and their development “reflect and constitute the cultural and social world that members of a society share and internalize” (Greif 2006, p.380). An economy and its characteristics need to be examined in a broader perspective, including historical developments.

6-2 Informal and formal institutions R Informal institutions are norms, customs and traditions in a society. R Formal institutions are the constitution, laws and economic rules in a society (rules of the game).

Figure 6-1 shows that the informal institutions (social embeddedness), representing the norms, customs, mores, traditions and so on, are located at the first level of the model. Informal institutions have historical origins as “they come from socially transmitted information and are part of the heritage that we call culture” (North 1990, p.37). Resulting from their origin they will change very slowly on the order of centuries or millennia (Williamson 1998). Commons (1961 [1934]) indicates that informal institutions are obtained through learning and explains their development as follows64:

64 “Concerns” refers to “going concerns”, which “in most general terms, […] is an organization of coordinated activity; it is collective behavior with a common purpose, and a collective will, governed by common working rules. Going concerns, as units of organization, occur in all phases of social life” (Commons et al. 1950, p.355).

109

121 Chapter 6 Institutional model for M&A

“[I]ndividuals begin as babies. They learn the custom of language, of cooperation with other individuals, of working towards common ends, of negotiating to eliminate conflicts of interests, of subordination to the working rules of the many concerns of which they are members. They meet each other […] as prepared more or less by habit, induced by the pressure of custom, to engage in those highly artificial transactions created by the collective will. […] Instead of isolated individuals in a state of nature they are always participants in transactions, member of a concern in which they come and go, citizens of an institution that lived before them and will live after them” (pp.73-74). North (1990) indicates that informal institutions (norms of behaviour) are very important in primitive societies with small communities: “[U]nder these conditions, it simply pays to live up to agreements. In such a world, the measured costs of transacting are very low because of a dense social network of interaction. Cheating, shirking, opportunism, all problems of modern industrial organization, are limited or indeed absent because they do not pay. Norms of behavior determine exchange and formal contracting does not exist” (p.55). As these societies become more complex, laws and regulations originate from these informal institutions. It is the ‘justice’ within the norms of behaviour that is formalized in the laws and regulations in a society.

Formal institutions - The second level represents the formal institutions and consists of the rules of the game, formed in the polity, judiciary, and bureaucracy. The main difference with the informal institutions is that the rules of the game can be enforced through the judiciary system. Formal institutions only come into existence when a society has grown more complex and the informal institutions become too difficult to enforce. Therefore, the origin of formal institutions lies in the originally accepted informal institutions within a society. “[T]he move, lengthy and uneven, from unwritten traditions and customs to written laws has been unidirectional as we have moved from less to more complex societies and is clearly related to the increasing specialization and division of labor associated with more complex societies” (North 1990, p.46). Formal institutions consist of the constitution, laws, and regulations. Laws define property rights, entailing the rights over the use and the income to be derived from property and the ability to alienate an asset or a resource (Eggertson 1990). The theory of property rights emphasizes the importance of clear definition and enforcement of these rights so efficient outcomes will be attained. Rule bargaining by various organizations takes place at this level. Laws and regulations are established and bargained over by organizations in society, and “reflect the bargaining strength of contractors, trade unions, and others in the political market” (North 1990, p.63). North (1990) also argues that “[I]n developed countries, effective judicial systems include well-specified bodies of law and agents such as lawyers, arbitrators, and mediators, and one has some confidence that the merits of a case rather than private payoffs will influence outcomes” (p.55). Enforcement of rules should be effective and it is important for transacting parties that by signing a contract both agree to live up to some set of standards. “[T]he property rights of an actor are embodied both in formal rules and in social norms and customs, and their economic relevance depends on how well the rights are recognized and enforced by other members of society” (Eggertson 1990, p.7).

110

122 Section 6-3 Governance structure and M&A activity

6-3 Governance structure and M&A activity R M&A activity is constrained by all other boxes and influences governance structure and individual actors. R Our model opens the neoclassical black box of governance structure, identifying two layers: the company and corporate groups. R According to transaction cost economics, transactions are aligned to alternative modes of governance in order to economize transaction costs.

Level three is where the play of the game takes place, governance structures are formed, and M&A activity is initiated. This level is constrained and influenced by informal institutions and formal institutions. We have placed solid bi-directional arrows between governance structure, M&A activity, and individual actors, because they have an important mutual influence over each other. M&A activity results from (strategic) considerations of individual actors, such as, for example, a company’s management, its shareholders, or a hostile raider. The governance structure of a company will have an important impact on these considerations and wether they will succeed. At the same time, M&A activity affects the governance structure, and can have an impact on the behaviour of individual actors in relation to stock ownership and M&A activity. Informal institutions have an important influence on decisions to engage in mergers, greenmail, and hostile tender offers. Based on norms of behaviour, individual actors can decide whether, and how, they engage in M&A activity. As will be explained in chapter 8, a company’s management, shareholders, and creditors are likely to be important in this process. The shareholders and creditors monitor and control the management through ‘corporate governance’. The term ‘corporate governance’ generally indicates the control of shareholders on management’s decisions through the board of directors. In the M&A process, also taking account of their role in a bankruptcy procedure, the creditors (can) play an important role which needs to be considered. In case of M&A, the role of shareholders is not only that of monitoring and control. Shareholders can approve or disapprove a merger or tender offer proposal, and with their ownership in a company they can also initiate or prevent a deal. 65 In this respect it is important to understand the motivation why shareholders decide to hold shares in a company. With the model it is also possible to examine the influence of other actors. Actors can be employees, government officials, bureaucrats, labour union members, private households, foreign investors, foreign companies and so on. As stakeholders, each actor (can) have a direct influence on M&A activity. In relation to the governance structure, informal institutions can influence (i) financing decisions, (ii) corporate governance, and (iii) formation of corporate groups. Regarding the financing decision, as a result of certain norms or customs, management might prefer a particular financing method over another. Informal institutions can also influence corporate governance when providing actors incentives, other than economic, to become shareholder of a company. The strength and cohesiveness of corporate groups in a society result from informal institutions as well. With the model it is also possible to investigate how formal institutions influence modes of governance, financing decisions, and M&A activity. In relation to the M&A activity and

65 Please refer to Chapter 8 for a more detailed description of the merger process.

111

123 Chapter 6 Institutional model for M&A

corporate governance, we can look into laws and regulations that influence ownership structures, mergers, and hostile takeover activity. Regarding the financing decision we can examine the laws and regulations related to debt and equity financing. Changes in ownership and corporate governance of companies can result from new financing possibilities, new financial institutions, and/or new investors. Regarding merger and hostile takeover activity, these can be influenced by: “[M]ore than a dozen separate forces drive takeover activity, including such factors as deregulation, synergies, economies of scale and scope, taxes, the level of managerial competence, and increasing globalization” (Jensen 1988). Other factors are problems and/or opportunities caused by alterations in economic conditions, efficiency conditions, and/or technology. On a macro economic level they can be related to industrial development, state policy, government regulations, or (international) competition.

6-3-1 Transaction cost economics As previously mentioned, governance structure in our model consists of two layers. At the higher layer we have positioned the corporate groups: horizontal keiretsu, vertical keiretsu, and trade association. At the lower layer the company is located and, according to transaction cost economics (“TCE”), we investigate the mode of governance in vertical relationships between companies. In New Institutional Economics the focus is on institutional arrangements, in particular the relation of transactions with different types of contracts and organizations. Institutional arrangements, also referred to as ‘modes of governance’, are governance structures such as companies, markets, hybrids and bureaus that coordinate economic transactions. TCE operates on this level and “invokes the discriminating alignment hypothesis, according to which transactions, which differ in their attributes, are aligned with governance structures, which differ in their cost and competence, so as to effect a (mainly) transaction cost economizing result” (Williamson 1998, p.37). The explanandum of NIE is “the governance structure to be defined as the institutional matrix within which transactions are negotiated, co-ordinated and executed” (Groenewegen and Vromen 1997, p.36). Informal and formal institutions, technology, and individuals’ preferences are taken as given in the explanation of institutional arrangements. Individual agents are opportunistic and have exogenous preferences and bounded rationality. They are “guided by considerations of self-interest to make allowance for strategic behaviour. This involves self-interest seeking with guile and has profound implications for choosing between alternative contractual relationships” (Williamson 1975, p.22). NIE is designed for specific research questions (optimalisation under constraints) and assumes that “the social institutions and norms, and changes to them, arise as the (intended and unintended) result of the decisions and actions of individuals” (Rutherford 1994, p.32). Institutions are the result of interaction between individuals. The individual is the only real actor and more important than the social whole. Individuals give society its aims and interest as society cannot have aims and interests of its own. NIE analyzes property rights at level two and the governance structure at level three while treating the informal and formal institutions as exogenous variables. Institutional change in NIE is advanced by examining changes in contracts and institutional arrangements, caused by exogenous changes in informal institutions, formal institutions and changes of technology. When there is no ‘Pareto’-optimum, the wealth-maximizing or

112

124 Section 6-3 Governance structure and M&A activity

cost-minimizing individual actors will renegotiate organizational structures and change the institutional arrangements.

The transaction cost economics forms the basis for NIE’s perspective on firms, markets, and property rights. The unit of analysis in transaction cost economics is the transaction. Commons (1950) proposes that transactions occur between “exchangers” that make “offers and bids to each other for the physical delivery of commodities […] these "exchangers" are sellers for money and buyers with money, and the accurate name of their dealings is alienation and acquisition of the two ownerships. This ownership is "property," meaning the "rights" of property. They are alienating and acquiring property rights in a commodity” (p.49). He also argues that the transaction “must contain in itself the three principles of conflict, mutuality, and order” (Commons 1932, p.4). Transactions are costly to identify and execute, and institutional arrangements arise in the search of economizing these costs. Coase (1937) argues that these costs are the reason why it might be more profitable to establish a company. He introduces the concept of transaction cost into the explanation of the firm; cost of organizing a transaction within the company need to be compared with the cost of using the price mechanism. Williamson (1998) explains that transaction cost economics concerns itself with the alignment of transactions to alternative modes of governance in order to economize transaction costs. Alignment is influenced by the behavioural attributes of human agents and the attributes of transactions. The behavioural attributes of human agents are bounded rationality and opportunism; bounded rationality results in incomplete contracts, and opportunism implies that a contract as a mere promise without credible commitments is not self-enforcing. The attributes of transactions are identified as (i) the frequency of the transaction, (ii) the uncertainty (disturbances), and (iii) the asset specificity of transactions (Williamson 1998, p.36). Frequency of the transaction relates to the information a party obtains when it is frequently involved in a transaction. This party will have more and ‘true’ information, which is not available to the party infrequently involved in the transaction. Uncertainty refers to the fact that we do not live in a stationary world and everything is subject to change. Asset specificity is most important in the transaction cost analysis and refers to transaction-specific assets of the supplier. These can take a variety of forms, such as physical assets specificity, human assets specificity, site specificity or dedicated assets. These assets have a contractual hazard (risk) because they lose value when the transactions are prematurely terminated. The buyer and supplier can set up safeguards for the contractual hazards such as a penalty for premature termination. As mentioned above, although frequency and uncertainty play a role, transaction cost economics focuses on the asset-specificity of transactions. Figure 6-2 shows the relationship between asset-specific transactions and governance structures. It indicates the influence of special purpose technology of the supplier (k > 0), general purpose technology of the supplier (k = 0), safeguards provided by the buyer (s > 0) and no safeguards provided by the buyer (s = 0). The basic alignment is that transactions without specialized investments and with a lot of competition between suppliers will be organized in “Ideal” markets. As contractual hazards rise, transactions will move to the hazard and hybrid structures. Hazard is the structure in which a supplier has made specialized investments but does not have any safeguards. In a Hybrid structure the supplier makes specialized investments and sets up contractual safeguards. The transaction

113

125 Chapter 6 Institutional model for M&A

cost theory predicts that higher contractual hazards will eventually lead to vertical integration of the supplier by the buyer company. This is termed as Firm (hierarchy) in which transactions are organized under unified ownership.

A ("Ideal market") k=0 B(Hazard) s=0 k>0 C(Hybrid) market safeguard s>0 administrative D(Firm)

Figure 6-2 Simple contracting schema Source Williamson (1998)

This process is the so-called “private ordering” in which governance “is the means by which order is accomplished in a relation in which potential conflict threatens to undo or upset opportunities to realize mutual gain” (Williamson 1998, p.37). It should be mentioned that complex forms of organization will be disadvantageous when contractual hazards are low, because hierarchies incur costs resulting from added bureaucracy.

6-4 Actors R Our model opens the black box of actors and includes shared cognition, individual mental maps, shared mental maps, and motivation in the analysis. R Shared cognition comprises all information that is commonly known by actors such as rules, beliefs, and norms inherited from the past and represent the preferences and conscience of individuals. R Mental maps are internal mechanisms and, as a result of (cultural) learning and experience, guide choice making. R Motivation represents the actors’ behavioural beliefs, internalized beliefs and internalized norms, motivating individuals to follow particular social rules or not. R Groups within a society are individuals with the same cultural background and experience, resulting in a shared mental model.

At level four the individual actors are located. They are constrained by the informal and formal institutions and, as discussed in the previous section, influence governance structure and M&A activity through behavioural attributes. Simultaneously, individual actors are influenced by governance structures and M&A activity. In our model we look into shared cognition, individual mental maps, shared mental maps, and motivation in our analysis of individual actors. Shared cognition comprises all information that is commonly known by actors such as rules, beliefs and norms inherited from the past, and represent the preferences and conscience of individuals. Although actors are aware of informal and formal institutions, they do not necessarily act accordingly. An individual actor has to be motivated to follow particular rules, and his/her motivation derives from belief in the rules and the actor’s preferences and conscience. Institutional analysis needs to focus on the

114

126 Section 6-4 Actors

question why some rules are followed and others are not. Greif (2006) mentions that “[A]t the center of analysis is the motivation to follow rules - and consequently beliefs and norms. It highlights the need to study rules and motivation to follow them in an integrated manner” (p.39). An important element in our model is the mental map that consists of habits, routines, cultural learning and experience. The mental map belongs to shared cognition and Denzau and North (1994) argue that individuals with the same cultural background and experience have the same mental models. We extend this argument to groups within a society and argue that members of a particular group in a society, with the same background, have shared mental models. As each individual also has an individual mental map, the degree to which an individual acts according to a shared mental map depends on the degree the individual belongs to a particular group.

Motivation - Greif (2006) defines an institution as “a system of rules, beliefs, norms, and organizations that together generate a regularity of (social) behavior” (p.30). Regularity of behaviour implies the behaviour that (most) individuals follow, or are expected to follow in a given situation. The regularity helps an individual in choosing his own behaviour based on the expected behaviour of others. Regarding the interaction between the individual and society, there are situations in which individuals cannot control an institution that influences their behaviour (institutions shape behaviour), and situations where the institution is an outcome of individuals’ actions. This dual meaning describes institutions as “systems of factors that are social in being man-made, nonphysical, and exogenous to each individual whose behavior they influence” (Greif 2006, p.44). Our analysis of the individual actor begins with the informal and formal institutions, the institutions that are commonly known and (i) create shared cognition, (ii) provide information, (iii) coordinate information, (iv) coordinate behaviour, and (v) indicate morally appropriate and socially acceptable behaviour. Shared cognition has the most important influence on behaviour because it represents the preferences and conscience of individuals. Motivation represents actors’ behavioural beliefs, internalized beliefs and internalized norms, and motivates individuals to follow particular social rules or not. Behavioural beliefs concern behaviour of others, internalized beliefs with regard to the relation between actions and outcomes, and the world around the individual. Internalized norms are “socially constructed behavioral standards that have been incorporated in one's superego (conscience)” (Greif 2006, p.37). Greif (2006) indicates that the “[M]otivation provided by beliefs and norms exogenous to each individual whose behavior they influence is the linchpin of institutions, as it mediates between the environment and behavior” (p.45).

Mental models - Denzau and North (1994) indicate that, in order to understand individuals’ decision making, it is necessary to include their mental models in the analysis. Individuals construct mental models “to make sense out of the world around them, the ideologies that evolve from such constructions, and the institutions that develop in a society to order interpersonal relationships” (p.4). The mental models are internal mechanisms and, as a result of (cultural) learning and experience, guide choice making. A theoretical framework therefore needs to include the interaction between informal institutions, formal institutions and actors.

115

127 Chapter 6 Institutional model for M&A

In this respect we would like to refer to Aristotle’s ancient Greek tri-partite conception of the soul. According to this conception, the soul should be thought of as a map of human capacities that is made out of three components. The soul has a vegetative component, an appetitive component, and a calculative component. The vegetative component is irrational and responsible for functions such as growth and nutrition (common to all species). The appetitive component has both an irrational and rational part and is responsible for our emotions and desires. It is irrational because emotions and desires can lead our behaviour (like animals), and rational in that (only) humans have the ability to control desires by reason. Proper control of our desires is called moral virtue and is formed in the vegetative and appetitive component of the soul. The calculative component of the soul is purely rational and responsible for the human ability to think and reason logically, evaluate, etc. Intellectual virtue stands for mastering these calculative abilities. It should be stressed that the rationality of the appetitive component and the calculative component is not similar. Only the calculative component stands for the actual capability of theoretical reasoning and practical thinking. The appetitive component is concerned with passions and actions and cannot itself reason, but is capable of following reason. Aristotle (350BC) shows that the habits that are developed during childhood are only the first step in the development of the soul. When habits are formed and we have the ability to reason, it becomes necessary to obtain practical wisdom in order to make rational choices. Rational choice is the result of humans’ ability to reason and evaluate instead of following uncontrolled desires and emotions. Aristotle observes that choice is related to things that are in our power, and to means rather than ends. Choice is different than a wish, because a wish is related to an end and can be for impossible things. Choice is not an opinion either; we can have an opinion about various things, including things that are beyond our power. Further, whereas an opinion can be false or true, a choice is either good or bad. Rational choice is the produce of reasoning instead of following uncontrolled desires and emotions arising from the appetitive component of the soul. That someone with great wisdom is able to make irrational choices results from the person’s inability to fully control the desires and emotions. We argue that the ‘virtuous behaviour’ with which Aristotle deals in his Nicomachean Ethics also applies to decisions of economic agents. Aristotle argues that it is ‘happiness’ or ‘living well’ that is chosen rationally. Happiness is a final state, desirable for itself and can be achieved through attaining wealth, honour, friendship, and/or pleasure. Thus, the difference in the general concept for economic agents is that wealth is not desirable for its own sake but for the sake of happiness. Depending on the individual’s rational choice, happiness can also be achieved with honour, friendship, pleasure, or a combination of these. This reasoning can also be seen in the work by Eggertson (1996), who explains that rational choice is often misunderstood in social studies that state that rational choice is culturally defined. He asserts that preferences and informal institutions are different among cultures and result in different rational choices. “[T]he rational choice approach only involves the assumption that individuals act consistently with their preferences when faced with alternative opportunity sets […] Preferences and institutional arrangement may differ among cultures, but it is not helpful, as far as we can see, to visualize choice as being a different phenomenon from one culture to another” (Eggertson 1996, p.18). Eggertson suggests making the rational choice richer by introducing altruism as a norm.

116

128 Section 6-4 Actors

Shared mental maps - The shared mental maps box reflects ‘groups’ of individuals with shared ideas and thoughts. By using ‘groups’ as an institutional element, we can consider various relations of actors that influence motivation and behaviour. We follow Hayek (1967) who argues that: “the properties of the individuals which are significant for the existence and preservation of the group, and through this also for the existence and preservation of the individuals themselves, have been shaped by the selection of those from the individuals living in groups which at each stage of the evolution of the group tended to act according to such rules has made the group more efficient” (p.72). Individuals influence the group’s rules of conduct and the group influences the individuals’ rules of conduct. Hayek (1967) explains how the group’s rules of conduct influence the individual’s rules of conduct as follows: “if behaviour according to the rules serves a sort of mark of recognition of membership of the group. If deviant behaviour results in no-acceptance by the other members of the group, and observance of the rules is a condition of successful co-operation with them, an effective pressure for preservation of an established set of rules will be maintained” (p.78). Rules of conduct influence the individual through the process of belonging to the group, but also by setting boundaries to the individual’s behaviour. Groups can have a structure in which in-group status is incorporated and relations of the members are important, but most important in a group is a set of norms with a distinct character of "we-ness". This distinct character will give rise to norms relating to out-groups, and reflects the behaviour and attitudes of individual members in interactions within the group and with out-groups (Sherif and Sherif 1953, p.230). Groups are formed in each society and their strength and cohesiveness depends largely on the informal institutions within the society. Sherif and Sherif (1953) explain that behaviour of individuals is different when they are alone or part of a group and argues: “[I]f characteristics prevalent among individuals of a group are products of group relationships and interactions, and not some essence of individual members, then the study of group differences must begin with analysis of these group relationships and interactions, and not with individuals in isolation. More specifically, it is difficult to foresee any value to the study of national character unless the implied group differences are related in a functional way to differences in social structure, stratification, degrees of differentiation, mode of life, and central values within the national group, and within its subgroups”(p.31). Sherif and Sherif (1953) also mention that the norms and values within a group need not be similar to those with regard to out-groups. Norms within the in-group can be democratic and cooperative, and not corresponding with the norms toward out-groups. In fact, “[D]epending on the nature of intergroup relations, at times the greater degree of solidarity and coöperativeness within the in-group may mean more effective friction with out- groups” (Sherif and Sherif 1953, p.209). In our model groups entail not only organizations formed by actors with similar background/nature, such as companies, courts or governmental agencies. Groups can also take shape with actors from a different background, e.g. managers and creditors, managers and shareholders, or managers and government officials. Further, it is possible that actors are part of more than one group. Through the relations of the individuals, inter-company groups are also located at his level, such as relations between two or more companies

117

129 Chapter 6 Institutional model for M&A

(managers), and a relation of a company with financial institutions (manager with shareholders and creditors). Each of these groups will influence the behaviour of the actor through the connection with the motivation box. By adding the individual mental model and shared mental model in the ‘actor’ box, our model provides an explanation for the difference between collectivist societies and individualistic societies. Motivation of the actor is influenced by the group it belongs to through the shared mental model, and herewith the group functions as an institutional element.

Groups - Greif (2006) distinguishes two types of societies; the collectivist society and individualistic society. The difference between the two societies lies in the cultural beliefs, the “shared ideas and thoughts that govern interactions among individuals and between them, their gods and other groups” (p.269). The cultural beliefs are attributes of individuals and determine societal organizations and influence institutional development. There is a fundamental asymmetry between beliefs inherited from the past and alternatives, and the beliefs from the past affect decisions. Greif (2006) explains that in an individualistic society the transactions are between actors from different groups. The actors are influenced only by their own motivation to follow social rules that lead to institutionalized rules and institutions. In a collectivist society, on the other hand, “each individual interacts socially and economically mainly with members of a particular religious, ethnic, or familial group” (p.269). In the collectivist society the actor is not only influenced by his own motivation but also by the shared beliefs and ideas within the group. The individuals have the same cultural beliefs, a long-standing relationship in which knowledge of past behaviour is available, and they belong to information-sharing networks. These networks are coordinating mechanisms and institutions that guide behaviour of its members towards each other and non-members. The network is used for information gathering and information transmission, which makes punishment of improper behaviour by a member or non- member self-enforcing. If a member’s behaviour is improper, based on customs and tradition, this information will be collected by the network and transmitted to other members, making collective punishment credible. All members know that collective punishment is possible and therefore refrain from improper behaviour. The result is that “[W]ithin these groups, contract enforcement is achieved through informal economic and social institutions. Little cooperation exists between members of different groups, but members of collectivist societies feel involved in the lives of other members of their group” (p.269). “[I]n individualistic societies, the social structure is “integrated” in the sense that economic transactions are conducted among people from different groups, and individuals frequently shift from one group to another” (p.269). Individuals do not have a long- standing relationship with each other and there is no knowledge of each other’s past behaviour. Each transaction is an individualistic, bilateral equilibrium that discourages investments in information gathering. There are no methods in which individuals can punish each other for improper behaviour and an external mechanism is needed. In the individualistic society “[C]ontract enforcement is achieved mainly through specialized organizations, such as courts. Self-reliance is highly valued.” (p.269)

118

130 Section 6-5 Institutional change

6-5 Institutional change R Informal institutions cannot be changed through an exogenous shock; changes are path dependent and occur in a matter of centuries or millennia. R Formal institutions can be changed upon initiative of a third party or through external pressure and are not necessarily path dependent. R In our model, institutional change is a historical process and needs to incorporate historical development of institutions and motivation of actors.

The most important difference between informal institutions and formal institutions is how these change and influence behaviour and decision making of actors. Informal institutions cannot be changed through an exogenous shock, and Williamson (1998) indicates that changes in informal institutions occur in a matter of centuries or millennia. The change in informal institutions will primarily derive from external influences such as new information or knowledge. Changes in informal institutions will follow a path dependent evolution, because they are present within individuals and transferred from one generation to the next. Further, the level of the external influences determines the way informal institutions change. If we suppose that a certain society is isolated and not exposed to any external influences, it is likely that informal institutions evolve slowly and it might indeed take millennia. Without new ideas or beliefs, the informal institutions that are transferred from one generation to the next will not change to a large degree. In an open society on the other hand, with a lot of external influences, it is unlikely that it takes millennia for informal institutions to change. We might have the same norms and values of our parents but our habits and customs will be influenced by new ideas and therefore be different. Also, when social structures consist of a large number of individuals, the influence from outside will be relatively small and change will take time. In contrast, formal institutions can be changed upon initiative of a third party or through external pressure. The pressures on formal institutions and economic development are considerably stronger than on informal institutions and can result from (i) foreign pressure, (ii) industrial development, (iii) changes in domestic and international political-economic conditions, or (iv) state policy. Changes in formal institutions are therefore not necessarily path dependent. Regarding the changes in formal institutions, the influence on the governance structure can be important. Figure 6-1 indicates that the formal institutions constrain the governance structure and the latter might be affected directly by the changes. The governance structure might however also change as indicated by North (1990), who explains that the difference in the rate of change for informal and formal institutions can cause tension: “the fact that the informal constraints that are culturally derived will not change immediately in reaction to changes in the formal rules. As a result the tension between altered formal rules and the persisting informal constraints produces outcomes that have important implications for the way economies change”(p.45). North (2005) mentions about path dependence that choices “are constrained by the heritage of institutions accumulated from the past” (p.51). He argues that path dependence means that “the institutions that have accumulated give rise to organizations whose survival depends on the perpetuation of those institutions and which hence will devote resources to preventing any alteration that threatens their survival” (pp.51-52).

119

131 Chapter 6 Institutional model for M&A

North (2005) also indicates that change will typically be incremental because “large-scale change will create too many opponents among existing organizations that will be harmed and therefore oppose such change. … Path dependence will occur because the direction of the incremental institutional change will be broadly consistent with the existing institutional matrix … and will be governed by the kinds of knowledge and skills that the entrepreneurs and members of organizations have invested in”(p.62). According to Denzau and North (1994) individual actors’ preferences and mental maps are endogenous and, as we explained above, (cultural) learning and experience is important. Campbell (1997) explains that the search process for a new governance structure is influenced by interaction, interpretation and bricolage. Problem definition is a subjective interpretation by actors with different interpretive frames, facilitating different understandings and solutions to problems. “…, as new frameworks of interpretation and their underlying interactive structures are locked in or institutionalized through repetition they become more or less taken for granted and provide the apparently ‘natural’ scripts that enable actors to define their problems and interests in broadly similar ways” (p.21). The evolutionary change is also a result of the process through which actors dealt with past problems and/or opportunities. This is the process of “bricolage, whereby new institutions differ from but resemble old ones” (p.22). Campbell (1997) mentions in this respect technical bricolage, whereby already existing institutional principles are recombined to increase efficiency or reduce inefficiencies. Symbolic bricolage is the process to make solutions acceptable within a broad social environment, by accompanying them with existing cultural symbols of language and rhetoric devices. “To the extent that actors frame institutional solutions by drawing upon these already existing cultural artefacts, their innovations are less likely to represent total breaks with the symbolic past thereby leading to evolutionary rather than radical institutional change at the symbolic level” (Campbell 1997, p.23). Campbell and Lindberg (1991) also argue that change is path dependent. Change can result from problems and/or opportunities caused by alterations in economic conditions, efficiency conditions, and/or technology. The state can also initiate a transformation “in response to domestic or international political-economic conditions” (p.329). These factors will induce companies to search for a new governance structure, which will be constrained by the old governance structure, institutional environment, and relationships with other groups and the state. The search process towards a new governance structure within an industrial sector is described as follows: “[T]he centerpiece of such an evolutionary view of sectoral development is what we have summarized heuristically as the search – a process whereby sectoral actors pursue their interests within a variety of constraints and eventually select a new governance regime by combining in intended and unintended ways their individual strategies for coping with the dilemmas of production and exchange. Indeed, the concept of constrained selection captures the essence of the search process as it embodies the reciprocal relationship between actor and constraint, or strategy and structure” (p.328) Following Greif (2006), we argue that the fundamental asymmetry between institutional elements inherited from the past and technologically feasible alternatives is important. It implies that beliefs, norms, and organizations inherited from the past affect the subsequent institutions. The institutional elements from the past are the attributes of institutions and

120

132 Section 6-6 Concluding remarks

individuals, reflecting the social and cultural worlds individuals know, forming the default in institutional change. “[I]nstitutional elements reside in individuals’ memories, constitute their cognitive, are embodied in their preferences, and manifest themselves in organizations; they are what individuals bring with them when they face new situations” (Greif 2006, p.188). New situations can evolve “when an institution that governed a transaction is no longer self-enforcing, when it is perceived to be losing its self-enforcing characteristics; or when technological, organizational, and other changes bring about new transactions” (Greif 2006, p.385). As long as the relevant parameters are in the range within which individuals’ behaviour is self-enforcing and reproducing, an institution will not change. When an institution is no longer self-enforcing, due to exogenous parametric change, it will seize to prevail. Institutional change is a costly process, involving introduction of alternative models, which implicates the necessity to gain new information to develop the alternative model. Institutional change is therefore “characterized by institutional refinement, that is, attempts to reinforce failing institutions rather than create new ones” (Greif 2006, p.195). The existing model is commonly known, followed by all individuals and therefore difficult to change. Generally, it is best to follow the existing model even if better alternatives are available.

6-6 Concluding remarks This chapter presented our institutional model for M&A. The model consists of four levels and each level needs to be studied against a historical and context-specific background, as this is the information that individual actors in a society share and internalize. Located at level one of the model are the informal institutions such as norms, customs, and traditions in a society; informal institutions constrain the three other levels. At the second level we place the formal institutions; the constitution, laws, and economic rules in the society (rules of the game). Formal institutions constrain governance structure, M&A activity, and individual actors. Located at level three are the boxes ‘Governance structure’ and ‘M&A activity’, and at level four the box ‘Individual actors’. Because these boxes influence each other, and are influenced by each other, they are linked with bi-directional solid arrows. M&A activity results from (strategic) considerations of individual actors, such as, for example, a company’s management, its shareholders, or a hostile raider. The governance structure of a company will have an important impact on these considerations and wether they will succeed. At the same time, M&A activity affects the governance structure, and can have an impact on the behaviour of individual actors in relation to stock ownership and M&A activity. Level three is the outcome of the dynamic interactions between economic development, (foreign) competition, and changes in informal and formal institutions that govern motivation and behaviour of individual actors. Behaviour of individual actors is the result of the interaction with level three and constrained by informal and formal institutions. The neoclassical black boxes of governance structure and individual actors are opened in our model. The governance structure consists of two layers: at the lower layer the individual company with ownership structure and corporate governance structure, at the higher layer corporate groups, such as horizontal keiretsu, vertical keiretsu, and trade association. According to transaction cost economics, we investigate the mode of governance in vertical relationships between companies.

121

133 Chapter 6 Institutional model for M&A

The box of individual actors is made up of the layers shared cognition, individual mental maps and shared mental maps, and motivation. Shared cognition reflects the actors’ knowledge of informal and formal institutions, and the mental models are internal mechanisms that guide choice making, influenced by (cultural) learning and experience. Motivation is important in the model, because an individual’s behaviour is the result of social rules that are motivated to be followed. Being a member of a group, individuals with the same cultural background and experience, results in a shared mental model. The shared mental model influences the motivation of the actor, and therefore functions as an institutional element in the model. Institutional change of informal institutions occurs in a matter of centuries or millennia, whereas formal institutions can be changed upon initiative of a third party or through external pressure. In our model, institutional change is a historical process and needs to incorporate historical development of institutions and motivation of actors. Only changing formal institutions does not result in different behaviour because “it requires creating new systems of interrelated institutional elements that motivate, enable, and guide individuals to take particular actions” (Greif 2006, p.402).

The informal institutions were described in chapter 3 and consist of Shinto, Buddhism, Confucianism, and the ie system. Formal institutions are formed by the political development and the formal institutions related to financial markets and M&A activity. The two layers of governance structure were described in chapter 4 and 5: ‘company’ refers to ownership structure and corporate governance structure, and ‘corporate groups’ to zaibatsu, horizontal keiretsu, vertical keiretsu, and trade association. We have not explicitly identified ‘individual actors’ in the previous chapters, but the model indicates that they are influenced by the informal institutions, formal institutions, governance structure, and M&A activity. In the next chapter we use the model to investigate hostile takeovers. In chapter 9 we will subsequently examine the impact of some of these institutional elements on shareholder wealth creation at the announcement date of mergers.

122

134 7

Institutional approach to hostile takeovers

In this chapter we analyze hostile takeovers in post-war Japan according to our institutional model for M&A. In section 7-1 we first discuss the interpretation in previous research of some institutional elements. We explain how the horizontal keiretsu and Japan’s cultural characteristics are argued to be related to the absence of hostile takeovers, and how transaction cost economics explains the vertical keiretsu. In section 7-2 we look into these elements in relation to our institutional model. We provide an interpretation of trust and discuss how informal and formal institutions have influenced companies’ modes of governance and group formation; we look into the horizontal keiretsu, vertical keiretsu and trade association. Section 7-3 uses our new insights on the institutional elements to interpret the hostile takeovers in the post-war period and the hostile tender offer attempts in the period 1999-2007. Section 7-4 concludes the chapter.

123

135 Chapter 7 Institutional approach to hostile takeovers

7-1 Interpretations from previous research R The horizontal keiretsu were formed as a result of hostile takeovers in the 1950s and prevented hostile takeovers in the entire post-war period. R Hostile takeovers do not fit in with the Japanese culture. R Hybrid structures without contractual safeguards can be sustained in Japan because of a high level of trust, importance of reputation, and financial hostages.

In this section we show how according to previous research, the horizontal keiretsu and the specific Japanese culture have limited hostile takeovers. Next, we show how transaction cost economics explains the vertical keiretsu.

7-1-1 Horizontal keiretsu The argument that the horizontal keiretsu were used as a defense mechanism originates from the hostile takeovers that occurred in the 1950s. It is reasoned that these hostile takeover attempts led Japanese companies to set up horizontal keiretsu with cross- shareholdings in order to protect group companies. According to Morck and Nakamura (2003), the buying back of the shares of Youwa Property by former zaibatsu group companies in 1952 “triggered the realization by top executives that corporate raiders could be blocked by establishing sufficiently large crossholding among former zaibatsu firms” (p.78). They suggest that in the 1950s a market for corporate control took off in which Japanese companies tried to gain control over other companies through hostile takeovers, and that raiders extracted greenmail from unwilling target companies (Morck and Nakamura 2003, p.3). They state that during this period Japanese companies had an Anglo-American corporate governance, but that the acquaintance was short and companies with dispersedly held ownership disappeared by the end of the 1960s (Morck and Nakamura 2003). Japanese companies established cross-shareholdings because they “would then all be safe from hostile takeovers, the need to pay greenmail would disappear, and the company presidents would have secure tenure in their jobs” (Morck and Nakamura 2003, p.78). Sheard (1989, 1991) mentions that just before the end of the U.S. occupation in 1952, Japanese companies started to buy each others' shares in order to prevent hostile takeovers, resulting in a considerable increase in intercorporate share ownership between the former Mitsubishi, Mitsui and Sumitomo zaibatsu companies and banks between 1949 and 1951. Another impetus to the cross-shareholding was the fall in stock prices in 1965 when companies increased their cross-shareholdings to prevent U.S. style hostile takeovers. The cross-shareholdings of the keiretsu companies, functioning as anti-takeover barriers, subsequently prevented hostile takeovers in the entire post-war period (Sheard 1989, 1991; Aoki and Patrick 1994; Aoki and Sheard 1992; Nakatani 1984; Morck and Nakamura 1999, 2003). The cross-shareholding was an effective anti-takeover barrier because the company’s stock was held by group members and motivated by relationships rather than share price appreciation. The shareholders bought the stock of group companies to symbolize the group loyalty and business links, and share ownership was not motivated by speculation (Jacobson and Aaker 1993). In fact, Sheard (1989) and Kester (1991) argue that the cross-shareholding structure eliminated the need for an expensive external market for corporate control, which is strongly related to the stock price.

124

136 Section 7-1 Interpretations from previous research

As we discussed in chapter 6, in previous research the company’s cross-shareholdings is taken as an exogenous variable without an analysis of the historical and context-specific background. The argument only looks into the development in the post-war period, but does not include the motivation of actors to be(come) shareholder. This focus prevents the researcher to look into the motivation of actors that do not belong to the company’s shareholders. Below we show that it is important to examine the influence of (changes in) formal institutions in relation to informal institutions, and motivation of individual actors.

7-1-2 Culture Another reason frequently given for the lack of M&A activity, especially hostile takeovers, is that it does not fit in with the Japanese culture. Resulting from cultural characteristics such as lifetime employment (and internal promotion), restricted labour mobility, seniority-based wages, and company unions, Japanese companies prefer internal growth to external expansion (Pettway and Yamada 1986; Odagiri and Hase 1989; Abegglen 1983). The company is considered to be made up of employees, having a responsibility for their well-being and future. The sale of a company, as a community of employees, is regarded as the sale of people and deemed immoral. The management of the target company will not accept a hostile tender offer (Abegglen 1983; Odagiri 1992). According to Kester (1991), the buying and selling of companies is seen as “fleshpeddling”, and the sale of a company is regarded as an admission of failure by the seller. Pettway and Yamada (1986) indicate that Japanese companies are operated for the benefit of stakeholders in the following order: firstly the managers and employees, then the main bank, followed by customers and suppliers, and lastly for stockholders. Sato and Hoshino (1984) and Abegglen and Stalk (1985) indicate that managers and employees believe that the company belongs to them. Managers prioritize market share, and stock price is least important of nine corporate objectives. Odagiri and Hase (1989) and Odagiri (1992) point out that if Japanese companies engage in M&A activity, capital participations and partial acquisitions are preferred to mergers and complete takeovers. Mergers and complete takeovers introduce problems related to differences in labor practices and cultures in companies, and result in uneasiness and conflicts of interests. Overall, the Japanese cultural characteristics result in a negative attitude towards merger and hostile takeover activities. Previous research takes the informal institutions as an exogenous variable to explain the low level of hostile takeovers. We think it is important to distinguish the arguments that are put forward. The first argument concerns the motivation to attain corporate growth through internal expansion, the second relates to the cultural characteristics such as lifetime employment and seniority-based wage systems. Although it is argued that the cultural characteristics developed from the management’s motivation, a change in the cultural characteristics does not necessarily originate from a different motivation of managers. Also, the focus lies on the reason why hostile takeovers do not occur, rather than identification of the nature of hostile takeovers and the motivation of actors to engage in hostile takeovers. The specific Japanese culture does not explicitly identify the motivation of actors to act according to the ‘harmony’ characteristic. This is assumed as an exogenous variable and there is no distinction made between individual actors.

125

137 Chapter 7 Institutional approach to hostile takeovers

7-1-3 Transaction cost economics and vertical keiretsu In our discussion of the vertical keiretsu we look into the subsidiary system. Previous research has indicated that, depending on the attributes of the transaction, the relation between the core company and suppliers differs (see e.g. Asanuma 1994; Dyer 1996). Suppliers can be independent companies (dokuritsu kaisha) or related companies (kankei kaisha). An independent company supplies standardized products and its relationship with the prime contractor closely approximates a "market" relationship. The core contractor does not have financial or business links with the independent company. In the automotive industry the independent companies generally supply numerous automakers and sell on average 20% of their output to a given automaker (Dyer 1996). The affiliated company, on the other hand, supplies the core company with non- standardized products that make up for the majority of the supplier’s total output. The two companies have a close relationship and the core company usually has stock ownership of the affiliated company, transfers technology, and dispatches employees such as technicians and senior executives (Gerlach 1987, p.132; Cusumano 1985). The parent company can exert financial, managerial and technological control over the subcontractors and the two companies engage in joint problem solving, price reduction and quality improvement and long-term relationships. The transactions range from design, contract assembly and systems-components subassembly, and demand the subcontractor to make asset-specific investments that are not (easily) applicable to other customers/contracts. The governance structure between the affiliated company and the parent company is the hybrid governance structure. Transaction cost economics (“TCE”) predicts that the governance structure of transactions primarily depends on transaction-specific assets and safeguards for the transaction. According to the theory, the choice of the governance structure moves from “Ideal” markets to Hazard and Hybrid structures, and eventually to the Firm (hierarchy) structure. In the ‘Hybrid structure’ the supplier sets up contractual safeguards, and in the ‘Firm structure’ the supplier is organized under unified ownership. The governance structure ‘Firm’ is equal to the vertical merger as will be explained in section 8-2 (the buying company absorbs the supplier). In this sub-section we investigate the choice of Japanese companies to organize transactions with specialized investments in the Hazard/Hybrid structure instead of the Firm structure. The interesting aspect of the vertical keiretsu is that the supplier makes transaction- specific investments without safeguards in the form of elaborate explicit contracts to protect against the hazards of opportunism, as predicted by Williamson (1985). TCE theory would predict that these Japanese transactors must have developed stronger safeguards to protect against the hazards of opportunism. Asanuma (1988) explains that the legal contracts in the automobile and the electronic industries are short standard contracts that only outline the general obligations of the parties, and effectively act as a "constitution" for the relationship. Research on the vertical keiretsu has been concerned with the question why the hybrid governance structure between the supplier and the prime contractor could sustain without contractual safeguards. It is argued that instead of contractual safeguards, Japanese safeguards in the vertical keiretsu are formed by trust such as collective responsibility, loyalty, reciprocal obligations, or harmony (Dore 1983; Sako 1991; Hill 1995; Smitka 1991; Powell 1990; Dyer 1996), reputation (Fruin 1992), and financial hostages (Klein 1980).

126

138 Section 7-2 Corporate groups according to our model

Trust is said to have been critical in lowering the hazards of opportunism and minimizing transaction costs. Williamson (1996) argues that Japan, as a society with high trust, requires less vertical integration and a larger amount of activity can be organized under hybrid structures (p.317). Dyer (1996) finds evidence which “confirms that Japanese automakers have been more effective than U.S. automakers at getting suppliers to trust them […] Japanese suppliers were more likely to trust Japanese automakers to treat them fairly and more willing to make dedicated investments based on oral agreements” (p.661). Dore (1983) suggests that "opportunism may be a lesser danger in Japan because of the explicit encouragement and actual prevalence, in the Japanese economy of what one might call moralized trading relationships of mutual goodwill" (p.463), and that "the sentiments of friendship and the sense of diffuse personal obligation which accrue between individuals engaged in recurring contractual economic exchange" (Dore 1986, p.460). Reputation can also serve to control opportunism in Japan because "the immediate gains from opportunism in a regime where reputation counts must be traded off against future costs" (Williamson 1991, p.291). Dyer (1996) defines reputation as the extent to which suppliers feel that the prime contractor is fair and trustworthy in business relations. Fruin (1992) argues that the horizontal keiretsu with the presence of the main bank and supplier associations enforce this reputation mechanism in the Japanese automotive industry. Exploitation or cancellation of business with a supplier is rather avoided by the parent company because it can result in a negative reputation. Financial hostages refer to stock ownership of the parent company in the supplier. The parent company indicates its commitment to the business relation with the supplier through these investments. The supplier herewith convinced about the motivation of the core company will make the transaction-specific investments required by automakers. Williamson (1985) refers to these mutual commitments as an arrangement akin to an "exchange of hostages".

7-2 Corporate groups according to our model R The Japanese society is characterized by high particularistic trust and low general trust. R The horizontal keiretsu finds its origins in the ie system and developed from the pre- war zaibatsu; it evolved as a result of the new formal institutions implemented after WWII. R The vertical keiretsu developed after WWII and is characterized by a relationship in which the parent company and subsidiary are mutually dependent. R Trade associations originate from 1092AD and have been very important in disseminating new technology and information to companies in the same business sector and in influencing government policy.

In this section we analyze the institutional elements from an institutional perspective. We look into the historical and context-specific background of the economy and examine the institutional elements because they “reflect and constitute the cultural and social world that members of a society share and internalize” (Greif 2006, p.380). In our analysis we pay special attention to the question how institutional elements are influenced by changes in formal institutions. According to institutional economics, new formal institutions only affect institutions when actors are motivated to follow them. Before we turn to the institutional approach of the three groups we look briefly into the concept of trust.

127

139 Chapter 7 Institutional approach to hostile takeovers

7-2-1 Trust In previous research on relations between Japanese companies the concept of harmony and trust is often used, suggesting it is a straightforward concept. It is argued that these characteristics have prevented Japanese companies from (i) engaging in hostile takeovers and (ii) opportunism in relationships within the vertical keiretsu. We define trust in company relations as involving uncertainty about the behaviour of the transacting partner and the preference to deal with someone that one party can feel safe about not being cheated by the other party under conditions in which cheating is possible. Yamagashi and Yamagishi (1998) argue that Americans are more trustful than the Japanese. In their insightful discussion, they distinguish general from particularistic trust. General trust is trust of individuals in general, the trust between individuals that do not have a particular relationship with each other. Particularistic trust is trust between individuals that know each other well and have a close and stable relationship. Particularistic trust prevents individuals to cheat each other because both have a lot of information about the other (knowledge-based particularistic trust), and their close relationship allows them to control the actions of the other (relation-based particularistic trust). Yamagishi and Yamagishi (1998) mention that “[W]hether a society is characterized by high levels of general trust or particularistic trust depends on how densely networks of mutually committed relations are established in the society” (p.113). They relate this to the perception on the nature of human beings by individuals. This is in line with institutional economics because individual’s behaviour is influenced by the behaviour of others. They argue that people with a high level of general trust have a positive view of human nature and are less inclined to form committed relationships based on high particularistic trust. People with a low level of general trust, on the other hand, have a more negative view of human nature and tend to build committed and stable relationships with a small group of business partners. They explain that networks of mutually committed relations in a society reduce uncertainty in these relations and increase particularistic trust. These relations, on the other hand, reduce the overall level of general trust. In Japan the particularistic trust is high and the general trust is low, and “[T]he relative lack of general trust among the Japanese moves them toward establishing networks of mutually committed relations. A fear of strangers (i.e., a low level of general trust) motivates the Japanese toward establishing mutually committed relations with persons they feel they can trust” (Yamagishi and Yamagishi 1998, pp.113-114). There is also a causal relationship between particularistic trust and general trust. When the particularistic trust in a society is high, leading transactions to occur in mutually committed relations, general trust will not develop to a high degree. The relationship between the level of general trust and particularistic trust is thus bi-directional, termed as a ‘reinforcing feedback loop’ by Yamagishi and Yamagishi (1998). They conclude that the feedback leads to two “ideal typical” societies. The first type is a society in which general trust is low and uncertainty in the society is reduced by extensive networks of mutually committed relations. “[T]his is a society in which people do not need to be honest and trustworthy; the networks of mutually committed relations take care of potential problems of social uncertainty” (p.116). The second type is a society in which people have a high level of general trust, not requiring the development of extensive networks of relations and resulting in a low level of particularistic trust. The second type of society is closer to the U.S. society in which general trust is higher than in the Japanese, which is similar to the first type.

128

140 Section 7-2 Corporate groups according to our model

This is an interesting addition to the discussion of Greif (2006) on collectivist and individualistic societies. Greif indicates that in a collectivist society the groups have religious, ethnic or familial origins. The individuals in the group have the same cultural beliefs and their long-standing relationship provides them knowledge of past behaviour. The collectivist society is characterized by individuals’ high particularistic trust. As Yamagishi and Yamagishi (1998) argue, the main motivation for individuals to form a group in these societies is the lack of general trust. In contrast, in the individualistic societies economic transactions are conducted among people from different groups because the level of general trust is high (either trust in transaction partner or law enforcing institutions). This discussion indicates that the elements of the Constitution of Seventeen Articles of Shǀtoku Taishi that we discussed in section 3-1-2, need to be considered in concert. The Japanese society is, in general terms, not characterized by harmony (article I), tolerance (article X) or trust (article IX). These characteristics primarily describe the relations within groups (article XVII).

7-2-2 Horizontal keiretsu In sub-section 7-1-1 we indicated that previous research suggests that in the years after WWII a market for corporate control took off and corporate governance of companies had Anglo-American characteristics. It is argued that hereupon horizontal keiretsu groups were formed as a means to prevent companies from hostile takeovers and that dispersedly held ownership had disappeared by the end of the 1960s. This sub-section examines the horizontal keiretsu from an institutional economics perspective. Rather than simply linking one event to another, in casu hostile takeovers leading to horizontal keiretsu, our institutional model argues that motivation and context-specific circumstances need to be included in the analysis. We need to analyze the motivation to form horizontal groups, and why companies did not opt for an Anglo-American takeover market. Supreme Commander for the Allied Powers (“SCAP”) had introduced U.S. laws and regulations that allowed the development of a competitive stock market. Yet, it did not evolve. The argument that the horizontal groups were formed to prevent hostile takeovers pushes the question a step back. Of course, high ownership by friendly and stable shareholders that do not sell their shares prevents hostile takeovers. The question that needs to be answered, however, is why the group structure developed. In this sub-section we first discuss the horizontal keiretsu in light of the historical development of institutions, and in section 7-3 we show that the causal relationship of hostile takeovers and horizontal keiretsu can be questioned and needs further exploration. As we explained in chapter 6, new institutions result from new transactions brought about by technological, organizational and other changes. Institutional change is costly and individuals therefore tend to reinforce failing institutions in stead of creating new ones. According to institutional economics, historical development of institutions and the context-specific background have an important effect on new institutions. Our analysis therefore includes the cultural and social world that members of a society share and internalize. SCAP implemented two important legal amendments related to corporate ownership after the war: the first was the change in influence from the general shareholders’ meeting to the individual shareholder, and the second was the prohibition to form holding companies. The combined effect of the amendments was a considerable increase of the individual shareholder’s influence. Companies strongly objected because they thought the

129

141 Chapter 7 Institutional approach to hostile takeovers

increased power of the individual investor would be abused and result in greenmail and hostile tender offers (Röhl 2005; Blakemore and Yazawa 1953). So, we need to investigate how institutions changed as a result of the new formal institutions in relation with the existing rules, beliefs and norms. Basically, we attempt to gain insight in the motivation to follow the new formal rules and/or the existing beliefs and norms. According to our institutional model, it is necessary to investigate the motivation for companies to form post-war horizontal keiretsu. This motivation is influenced by the shared cognition of individuals, resulting from experience, knowledge and cultural norms and beliefs with origins in the preceding period(s). As we explained in chapter 4, the Mitsubishi, Mitsui and Sumitomo keiretsu originate from the pre-war zaibatsu. In order to understand the formation of the horizontal keiretsu group, it is therefore necessary to understand why and how the pre-war zaibatsu were formed. Further, the explanation of the corporate structure in the pre-war period needs to be retrieved in periods prior to the Meiji period (1868-1912). The pre-war zaibatsu evolved from the ie system that was present in agricultural communities and merchant companies during pre-Meiji Japan. In chapter 3 we explained that the house in the ie system was a highly independent economic unit and persisted over generations. In the succession of the leader of the house the continuity of the house was more important than the heir. A son who did not become the leader of the honke (the most senior house) could marry into another house or set up his own house, bunke (junior house). If he set up his own house, this was regarded as a branch of the main house, and through this system a wide group of houses with personal and financial linkages (dǀzoku) developed. Loyalty to the company was very important and necessary financial means were raised within the group of houses. Partnerships or anonymous investors did not play a role in corporate finance. Chapters 4 and 5 showed that the pre-war zaibatsu group development was similar to the ie system. Internal financing was most important and external finance not used until the 1930s. In the 1920s companies were hived off from the parent company, and a group was formed in which loyalty to the parent (holding) company and group companies was important. The discussion of succession of the leader of a company is also interesting in this respect. Both in the ie system and the zaibatsu group, the most important aspect was the continuity of the company and group. “[O]riginally founded as family businesses, the groups operated until their dissolution on the principle of absolute loyalty of the company managements to the regnant families” (Caves and Uekusa 1976, p.2). Figure 7-1 shows the transformation of the ownership structure in the Mitsubishi group during the pre-war and post-war periods. Based on our discussion in chapter 5, the group’s transformation can be divided according to three periods: (i) up to 1920s, (ii) the 1930s, and (iii) post-war. The organizational structure of the Mitsubishi zaibatsu up to the 1920s was simple. The holding company and Iwasaki family owned the shares of the main companies and financial institutions. The main companies and financial institutions, together with the holding company and family, had shareholdings in the other zaibatsu group companies. In the 1930s, due to the anti-zaibatsu pressures, the holding company and family started selling shares and a different ownership structure developed. Buyers of the shares were unrelated investors but also group companies, especially insurance companies. The result was that the holding company and family were not completely alone at the top of the pyramid anymore. Financial institutions started to hold shares of the group’s main companies. The holding company, financial institutions and main companies all had shareholdings in other related companies.

130

142 Section 7-2 Corporate groups according to our model

Up to 1920s Holding / Family

Main Companies & Financial institutions

Other Companies

1930s Holding / Family Financial institutions

Main Companies

Other Companies

Post-war Diamond (main) Financial institutions companies

Other Companies

Figure 7-1 Ownership transformation of Mitsubishi groups

In the post-war period it was not allowed to form pure holding companies and the new horizontal keiretsu system developed. The keiretsu diamond companies66 and the financial companies replaced the holding company. The diamond companies have shareholdings in the most important group companies and the financial institutions. The financial companies, particularly the main bank, took the role of the holding company with shareholdings in all group companies. The developments in the pre-war and post-war periods indicate that continuity was most important in the groups. Changes in the economic and regulatory environments caused transactions to change, but the enforcement of continuity of the group resulted in ownership structures that could sustain this. Regarding the post-war period, the regulatory changes resulted in the prohibition of the holding company and a stronger position of the individual shareholder, and horizontal keiretsu were formed to ensure the continuity of the pre-war group. The importance of the zaibatsu family can also be derived from the fact that a lot of group companies returned to their group logo when it was permitted again in 1952. The importance of the group can be traced back to the pre-Meiji period and corporate groups are likely to have been formed irrespective of the hostile takeovers in the 1950s. Next we try to clarify that the causal relationship between hostile takeovers in the 1950s and the formation of horizontal keiretsu can be questioned by explaining the chronological order of the events. The main difficulty in this discussion is that only in 1952 zaibatsu

66 Please refer to sub-section 5-3 for our discussion and definition of the “keiretsu diamond”.

131

143 Chapter 7 Institutional approach to hostile takeovers

companies and financial companies were allowed to hold shares of former zaibatsu group companies. As the hostile takeovers occurred in the beginning of the 1950s, this coincides with the laws allowing for cross-shareholdings. Does this mean, however, that the relationship is causal? Suzuki (1997) provides evidence that the zaibatsu companies tried to amass their own shares prior to the hostilities. He explains how Mitsui Real Estate coped with the transfer of shares by the Holding Company Liquidation Commission (“HCLC”) after the Mitsui zaibatsu dissolution. Mitsui Real Estate increased its capital from 100,000 shares at 5 million yen to 1,000,000 at 50 million yen and the newly issued shares were distributed by HCLC. A small part of the issued shares were allotted to company managers and employees (60 thousand shares) and the remaining 840 thousand shares were sold to the public at 400 yen. The management of Mitsui Real Estate made secret purchases of 280,000 shares (28% of outstanding shares) through Nomura Securities. Nomura Securities functioned as “owner” of the shares and the purchase of about 120 million yen was financed with funds from the Teikoku Bank, Chiyoda Bank, Joyo Bank and other financial institutions. Suzuki (1997) explains that this allowed Mitsui Real Estate “to gain control over its own stock. Other companies employed similar methods to effect protection against takeovers, and it is for this reason that the major securities houses are found listed as major stockholders in these companies around 1950” (p.78). Our discussion on the new company law that reduced the influence of the general shareholders’ meeting, together with the evidence from Suzuki on Mitsui Real Estate, indicate that the ownership structure of group companies were of great concern for former zaibatsu companies. We showed that the formation of keiretsu was not caused by the hostilities in the 1950s, but that they evolved as a consequence of new laws and regulations. The new formal institutions resulted in fear of the group’s discontinuity as they prevented control of companies through holding companies and created opportunities for greenmail. The lack of general trust had resulted in the formation of groups in the pre-Meiji and Meiji periods and also led to the formation of the horizontal keiretsu in the post-war period. This section showed that it is necessary to investigate various institutional elements to be able to explain and understand the creation of horizontal keiretsu, and in section 7-3 we investigate whether they prevented hostile takeovers in the post-war period.

7-2-3 Vertical keiretsu The development of the vertical keiretsu is different from the horizontal keiretsu as it finds its origins in the war-time period instead of the pre-war period. Although subcontracting existed in the pre-war period, it was considerably different from the post-war vertical keiretsu. Similar to our discussion of the horizontal keiretsu, comprehension of the pre-war subcontracting system is indispensable for understanding the vertical keiretsu. It provides insights in the development of the vertical keiretsu and indicates that the business chain was not based on feelings of friendship or trust. In fact, the vertical keiretsu developed because of the lack of both and was aimed at profitability of the parent company and subcontractor company (often subsidiary). In the pre-war period the relationship between the central company and the subcontractor had two characteristics. The first was that the central company tried to reduce its cost by forcing its subcontractors to accept a low price for products. The central company could coerce subcontractors to accept its demands because of its capital superiority and very strong position in the transaction. The weak position of the subcontractor was caused by the simple

132

144 Section 7-2 Corporate groups according to our model

nature of products, excessive competition between small and medium sized enterprises, and the lack of technology, marketing power and alternatives to sell its products. The parent company used the subcontractors because of the low price and also as a buffer of business cycles; in a business depression the parent would divert production from the subcontractor and produce the product in-house. During WWII, war-related companies had to expand their production rapidly but faced problems with increasing production capacity and product quality improvement. The military authorities took the initiative to solve these problems by requiring central companies to get involved in the production process of subsidiaries. It was acknowledged that production capacity and product quality could not be improved if the technological knowledge and quality of supplies remained the same. Induced by to the so-called ‘Co-operation Factory System’, central companies started extending technical assistance and leasing and selling machines to the subcontractors in order to solve the mentioned problems (during the war primarily in the metal industry). “[T]he subcontracting relations were no longer determined by the one-sided interest of the central enterprise, but by the higher standpoint of social productive power and this accelerated the progress of the medium-small enterprise” (Tasugi 1963, p.36). Immediately after the war, the subcontracting relations returned to the pre-war structure of a strong central company exploiting the subcontractors. Resulting from the economic development induced by the Korean War boom of 1951-52 and also the boom in 1955-56, the central companies tried to expand production as demand increased. Similar to the war-period, they faced difficulties increasing their production capacity and product quality. Furthermore, due to the self-centred attitude of the central companies in the previous period(s) subcontractors were reluctant to assist the central companies in their expansion activities. Subcontractors tried to select central companies that considered their position and, as they were crucial in the production process due to their lower cost than the parent company, the parent companies had to change their attitude towards the subcontractors. As competition between large companies intensified with economic development, and these companies wanted to secure supplies from subcontractors, the latter came to hold a stronger position in negotiations. The subcontractor’s position strengthened even more in the 1960s when accelerating product proliferation and short product life cycles began to characterize product markets. Flexibility was needed and central companies came to realize that the ability of subcontractors to accommodate the required supplies was crucial for the increasing product mix and variants, design changes, quick and frequent deliveries without incurring prohibitive costs of implementation. In the new approach, large companies delegated assembly of finished goods and subassembly or manufacture of system components to subcontractors. Parent companies directed their efforts to the growth of the subsidiary and associated suppliers by providing assistance in various ways such as training, dispatch of technicians or senior executives to work at the supplier on a temporary or permanent basis, et cetera. The subcontracting became strategically oriented and systematically executed. In the automotive industry, for example, the prime contractor began to produce large-volume products and the most important final products. The task of subcontractors changed from uncomplicated tasks in the pre-war period and the first years of the post-war period to assembling small-volume specific products and special-purpose vehicles in the 1960s. In this way the prime contractor was relieved from operational and administrative tasks that became increasingly more difficult and complex. The prime contractor could focus on consumer demand and the competitive market, while the subcontractors focused on

133

145 Chapter 7 Institutional approach to hostile takeovers

technological innovations and developments to shorten the product manufacturing time. The subcontractors converted into subsystems manufacturers and contract assemblers, according to their specialties. Another change in this period was the delegation of tasks such as testing and design to the subcontractors. The main aim of the subsidiary and parent company was to increase the competitiveness of the group and enhance their own profitability. Based on the preceding discussion we argue that the relation between the parent company and the subsidiary in the vertical keiretsu cannot be advanced with transaction cost economics. Transaction cost economics is applicable when the trading relationship between two independent trading parties is analyzed. In those situations the attributes of the transactions and opportunism are important. We have shown that in the vertical keiretsu, however, the parent company and subsidiary company have a mutual benefit in the transactions. The motivation of the two parties is economic and the agreements will be upheld as it affects the profitability of both companies. The fact that the relationship is based on economic motivations can also be deduced from the two-vendor practices of the parent company and replacement of a subsidiary, if supplies do not meet the necessary standards. This is also known by the subsidiaries that make efforts to maintain the required level of quality. The fact that subsidiaries in the vertical keiretsu are stable shows that, in principle, the two parties work together to improve product quality and profitability.

7-2-4 Trade associations Trade associations are another corporate group we consider to be important to include in the analysis of M&A in Japan. The first officially recorded trade organization in Japan materialized in 1092AD. A Kyoto shrine, as a patron, guaranteed a group of merchants exclusive rights of trade for specific products in its geographic area. This kind of group is named za and are “groups of artisans and merchants with closed membership (i.e., new merchants could not easily joint), official standing granted by a patron” (Schaede 2000, p.218). The za had a positive effect on the economy while providing the patron tax income, and functioned as self-regulation of trade in the absence of formal institutions. During the 15th century, when Chinese coins became a recognized currency in Japan 67 ,theza increased in importance. The recognition of Chinese coins as a valid currency led to imitations of low quality to come into circulation, increasing the risks of being cheated for merchants. “Merchant dishonesty was common, and almost expected, when transactions were not on a regular basis” (Sheldon 1958, p.41). To prevent being cheated, and to have the opportunity to hold someone responsible, merchants opted to avoid unknown merchants and trade with “members of the various za,sinceza membership was a long- term affiliation and bred interdependency” (Schaede 2000, p.219). The za evolved from a means to stimulate economic development by a patron to a self-regulation system independent of patronage. During the Tokugawa period trade expanded rapidly and grew interregional and the distribution system became more complex. Merchants shipped “products from the countryside into the cities, changing rice into money, running warehouses for the daimyo in the large cities, and providing credit for the daimyo” (Schaede 2000, p.224). In this new trading environment the trade associations became more sophisticated and merchants

67 Japan’s first coins were minted in 708 and copper. During the 10th century these Japanese coins were replaced by coins from China , the latter becoming a recognized currency in the 15th century.

134

146 Section 7-2 Corporate groups according to our model

established so-called nakama. Nakama are fixed-membership trade associations and its literal meaning is “among those who know each other”. Sheldon (1958) argues that nakama were primarily motivated by the “feeling of need for mutual protection and the feeling of group solidarity” (p.54). The nakama evolved into kabu nakama and functioned as monopolistic guilds that limited entry to the associations, colluded in price setting and other rule-setting (quality, and so on.). Members were not allowed to deal with outsiders, be that in their own trade or with outsiders in other product categories. Another favoured barrier to entry was to demand certain quality and other standards. Sheldon (1958) explains that “[K]abu has often been translated as "share," but this is not quite accurate, as ownership was hereditary and the nakama could restrict transfer. The only entrance to an established nakama was through inheritance, the purchase of kabu which had been "vacated" by bankruptcy or lack of an heir, and through the apprentice system, all subject to the approval of the nakama itself” (p.52). The nakama provided its members a reputation of creditworthiness towards other nakama, and collective action could protect its members from being cheated by members of other nakama. In this system the trustfulness of the nakama to outside nakama was most important and if a nakama-member did not act according to the rules, he could be expelled or forced to pay a penalty. “To make this deterrent even stronger, many nakama set up an “insurance fund” to which everyone had to contribute a sizeable amount. In the event that one member broke the rules, e.g. by undercutting prices, his contribution would be used to pay off those members who had incurred damage through his behaviour” (Schaede 2000, p.229). Given the long-term costs that distrust meant to a merchant, the nakama were quite successful in policing their members. Although various nakama developed during the Tokugawa period, the Bakufu had opposed them68 and only in 1721 all kabu nakama received official licensing. “Each kabu nakama was required to pay a small annual fee (myǀga-kin), and each new member was required to pay this fee to the Bakufu upon entrance.” Prior to 1721 only a limited number of kabu nakama had been officially recognized such as pawn shops (in Osaka, 1642; in Edo, 1692), second-hand dealers (Osaka, 1645; Edo, 1702), public bath houses (in Edo, 1650), and for peddlers and hairdressers (in Edo, 1659). Money-changers, unable to exclude non-members from entering the market, were officially recognized in 1679. Up to the Meiji period there were periods in which they were allowed or banned. Overall, the kabu nakama provided a system of self-regulation for merchants to cope with mistrust, inducing competition on product quality (and its improvements) rather than price (Sheldon 1958, p.57). The Meiji government abolished all kabu nakama in 1871 but, due to the absence of an institutional environment, new local organizations emerged and many resembled the pre- Meiji nakama.Thekabu nakama evolved in the trade associations as discussed in chapter 4. Formed by companies from the same industry the associations shared technical information, protected members’ privileges, established trade rules, and endorsed credit for members and controlled quality and prices. The associations were also important in their representation of the industry in discussions with politicians and, as such, could influence political decision-making. Cross-sectoral associations were also established to this aim, in particular to influence tax, trade, and investment policies. In 1917 the Japan Industrial

68 The Bakufu was particularly vigilant against any attempts to corner a market, and any collusion among merchants and officials to make big profits.

135

147 Chapter 7 Institutional approach to hostile takeovers

Club (Nippon Kogyo Kurabu), precursor of Nikkeiren, was established. The club was dominated by the zaibatsu companies and its members comprised of 185 large companies and primarily supported the growth of the heavy and chemical industries after its establishment. After 1922 this group was predominantly involved with labour issues and pressuring the government for economic reform. The Japan Economic Federation (Nippon Keizai Renmei), predecessor of Keidanren, was set up in 1922 and took over most of the direct political activities of the Japan Industrial Club. This group is characterized by the strong personal relations between politicians and managers of major companies. Two other cross-sectoral associations are the Japan Chamber of Commerce and Industry (Nippon Shoko Kaigisho), for small and medium-sized businesses, and the Keizai Doyukai,a discussion forum for younger executives (Lynn and McKeown 1988; Shaede 2000). We conclude that Japanese companies decided to arrange the predecessors of the trade association, za and kabu nakama, to prevent being cheated by unknown traders. The lack of a legal system resulted in strong relations of companies in the same industry. Similar to the discussion by Yamagishi and Yamagishi (1998), due to the low general trust in the society, groups were established in which particularistic trust was high. As trade associations have been present in Japan since 1092AD, we conclude it is an important institutional element and we examine its influence on hostile takeovers in the next section.

7-3 Hostile takeovers according to our model R Prior to 1999 there were no hostile tender offers, hostile takeovers were not aimed at gaining operational control over a company and should therefore be seen as greenmail attempts. R Vertical keiretsu and trade associations have an important effect on hostile tender offers.

In this section we take an institutional approach to hostile takeovers in post-war Japan. We already mentioned in chapter 1 and chapter 2 that it is difficult to make a distinction between greenmail and hostile tender offers, but there is an important difference in the motivation of the bidder. In tender offers the bidder publicly announces it wants to acquire a company (or a certain percentage of the outstanding shares) and states its motivation or intention for the acquisition. The public announcement allows the targeted company and/or other companies to act in response to the tender offer. As we explained in section 7-1, the horizontal keiretsu (with stable and cross- shareholders) and the Japanese culture are said to have prevented hostile takeovers in the post-war period. Odagiri (1992) comments on the relation between the stable shareholders and hostile tender offers as follows: “If, for instance, they are dissatisfied with the current management and offered a high premium on their shares, will they really decline the offer? It is difficult to answer this very interesting question, however, precisely because such cases have been rare” (p.105). In sub-section 7-2-2 we explained that it is unlikely that the hostile takeovers in the 1950s were the only, or main, reason for the formation of horizontal keiretsu. In this section we examine the commonly asserted relation between the horizontal keiretsu and absence of hostile takeovers. Is this assumed relationship valid or, resulting from the interpretation of the 1950s with reference to the horizontal keiretsu, a self-fulfilling prophecy? “[T]he self-fulfilling prophecy is, in the beginning, a false definition of the situation evoking a new behavior which makes the originally false

136

148 Section 7-3 Hostile takeovers according to our model

conception come true. The specious validity of the self-fulfilling prophecy perpetuates a reign of error. For the prophet will cite the actual course of events as proof that he was right from the very beginning” (Merton 1948, p.195). We begin our analysis with the question why hostile tender offers did not occur in Japan prior to 1999, and then examine the hostile tender offers in the period 1999-2007. Let us start by acknowledging that it is possible that hostile tender offers were not initiated because bidders thought that the attempt was deemed to fail as a result of the horizontal keiretsu. As such, it can be argued that the horizontal keiretsu defended companies from hostile tender offer attempts, and functioned as an implicit defense structure that never needed to be used. Below we show that this reasoning is unlikely and that, in fact, it is more probable that there were no hostile tender offer attempts because of other institutional elements. We need to consider the motivation of the bidder, whether the bidder was interested in taking control of the company or getting paid greenmail. As we explained in chapter 2, there is a strong relationship with a (failed) hostile tender offer and greenmail, which makes it difficult to separate the two. The combination of the hostilities that occurred and the formal institutions in the post-war period, however, suggest that the hostile takeovers were aimed at greenmail. Most important in this respect is that Japanese companies could not buy back their own shares prior to 2001. This implies that bidders that were interested in greenmail would aim their hostilities towards companies that belonged to groups. The group members would be able to buy back the shares from the greenmailer. Independent companies might not be able to find friendly companies willing to purchase its shares and the greenmailer would not be able to sell. Conform this argument, horizontal keiretsu were often targeted by greenmailers. As also suggested by Kester (1991), the stable shareholdings in the horizontal keiretsu groups were an advantage to greenmailers that had no intention of getting operational control of companies. We think that the hostile takeovers in the post-war period were nothing other than greenmail, and that the bidders did not have the intention to get control over a company. In fact, if a bidder really wanted to gain operational control over a company why did he not aim its activities on a company independent of a horizontal keiretsu? In 1987, for example, according to the Keiretsu no Kenkyǌ the member companies of the 6 horizontal keiretsu totaled 24 for Mitsui, 29 for Mitsubishi, 20 for Sumitomo, 29 for Fuyo, 44 for Sanwa and 47 for Ikkan. This makes a total of 193 companies related to a horizontal keiretsu compared to 1,724 companies listed on the first section of TSE. If companies were indeed interested in hostile takeovers, and horizontal keiretsu prevented these, it appears there were enough companies that were not part of a horizontal keiretsu and more likely to be a successful target. Yet, there were no hostile tender offer attempts prior to 1999. In order to understand why Japanese companies did not initiate hostile tender offers or hostile tender offer attempts, we think it is important to add two institutional elements in the analysis: the vertical keiretsu and the trade association. The vertical keiretsu prevents hostile tender offer attempts in three ways. The first important characteristic is that the Japanese company cannot be seen as a separate entity as for example U.S. companies. As we explained in section 7-2-3, the vertical keiretsu is a group of companies with interdependent relations in the production process. Rather than the production of products within the company, the Japanese company uses a wide subcontracting system that makes it a difficult target for a hostile tender offer. The second characteristic of the vertical keiretsu is that a parent company will be in great difficulty if a subsidiary would be taken

137

149 Chapter 7 Institutional approach to hostile takeovers

over by another company, because of the subsidiary’s role in the production process. The majority of its important subsidiaries is already fully owned, but if a hostile takeover is attempted on an important partially owned subsidiary, the parent company will intervene to secure its own production process. This important relationship between the parent company and the subsidiary can also be seen in the greenmail attempts on subsidiaries. The third aspect is the question whether it would be beneficial for a hostile bidder to try and takeover an unwilling subsidiary company. When a hostile tender offer is forcefully pursued against the will of a company’s management, which has good relations with its suppliers and customers, the successful bidder company will probably lose these good relations in the process. The other aspect that is important in preventing hostile tender offer attempts in Japan is the trade association. The discussion of horizontal keiretsu limits itself to the way a hostile tender offer can be prevented by the shareholdings of member companies. It, however, does not explain why a company that does not belong to the same horizontal keiretsu would not initiate a hostile tender offer attempt against a company from another horizontal keiretsu. As indicated in section 7-2, this is explained by the assumed Japanese cultural characteristics of high level of trust. However, as the level of general trust is low, we could expect that companies of different horizontal groups would engage in hostile tender offers. We argue that the reason they do not engage in this kind of activity could derive from the fact they belong to the same trade association. The trade associations originated to prevent being cheated by unknown traders and in these associations companies in the same industry participate and exchange information. They also discuss how to influence the government regarding new laws and regulations. As such, companies in the same industrial sector have close relationships, particularistic trust is high, and hostile tender offer attempts could hereby be prevented.

7-3-1 Hostile tender offer attempts Chapter 2 shows that prior to 1999 there had not been any hostile tender offers but that in the period 1999-July 2007 thirteen hostile tender offers were launched. Of these hostile tender offers one was aimed at attaining more than one third of the outstanding shares, three at obtaining more than one half of the outstanding shares, and nine to obtain full ownership. Two of the hostile tender offers were successful, two companies could fend off the hostile tender offer by increasing dividend payout, in four cases a white knight rescued the target, four target companies implemented a poison pill as a defensive measure, and one company issued a poison pill and set up white squires. In our discussion we only look into hostile tender offers and exclude the off-market purchase of NBS by Livedoor and the request to merge by Rakuten after acquiring 19.09% of TBS’s shares. These cases, especially the Livedoor purchase, were highly publicized and important in the sentiments towards hostile takeovers. They resulted in implementation of hostile takeover defense measures by many companies, but from an institutional perspective, the Livedoor case is an example of smart financial engineering within the boundaries of an immature legal system. As we showed in chapter 3, the law has been amended as a result of the Livedoor event to prevent companies purchasing more than one third of a company’s outstanding shares and circumventing the TOB regulations. The law was also amended after Don Quijote’s purchase of shares in serial purchases without notification. In order to understand whether cultural characteristics, horizontal keiretsu, vertical keiretsu or trade associations had an important effect on hostile tender offers, we analyze

138

150 Section 7-3 Hostile takeovers according to our model

the characteristics of the tender offers. We do this by answering the following questions: Which companies are targeted, and who are the actors that initiated the hostile tender offers? When looking at the targets of the hostile tender offers we see that these do not involve companies in a vertical keiretsu. Further, only one of the target companies, Shoei, belonged to a horizontal keiretsu with 60% of its shares owned by group-members. This is an important difference from the greenmail practices in the period prior to 1999 when horizontal keiretsu companies were primarily targeted. Whether the hostile tender offer of Shoei failed because of the group relations is doubtful. It is argued that the bidding price of 1,000 yen was far too low and the share price had increased to 1,480 yen after the bid. These circumstances did not require actions by a white knight and implementation of anti- takeover defenses was not necessary. The other target companies were independent companies and increased dividend, used a poison pill, or asked assistance from a white knight to prevent being taken over by the hostile bidder. Apart from the companies that implemented poison pills, it can be argued that the other failed tender offers had an effect; two companies increased dividend payout, and all white knights that prevented the hostile takeover belonged to the same industrial sector as the target company, implying an industrial restructuring. Regarding the white knights this might suggest that the relations not only prevented hostile tender offers in the period prior to 1999, but also functioned as a means to prevent companies falling into the hands of hostile bidders. Regarding the latter point, we need to address the hostile tender offer made by Oji Paper, the largest paper manufacturer of Japan, for Hokuetsu Paper, the sixth largest. Oji Paper and Hokuetsu Paper both belong to the Japan Federation of Printing Industries. This raises the question whether the influence of the trade association is changing. Oji Paper’s reason for the hostile tender offer was the necessity to build a company that is competitive in the international market. It explains that the two companies combined would create the number 5 paper company in the world. At this size the company would be better able to deal with the costs of oil, power and materials. According to a report by Hokuetsu Paper there have been continued discussions with Oji Paper regarding the enhancement of international competitiveness the last few years. During these discussions Oji Paper has (coercively) requested Hokuetsu Paper to stop planning the construction of a cutting edge facility several times. Oji Paper had not reached the level of technology to produce a similar facility yet. Because Hokuetsu Paper did not comply with the request, Oji Paper launched its hostile tender offer after the official announcement of the factory’s construction. This hostile tender offer was not only to the dislike of Hokuetsu paper. The Japan Federation of Printing Industries issued the following press release on August 8, 2006: “The market share of post-integration of Oji and Hokuetsu will likely violate the anti-monopoly law. This integration is damaging to the printing industry and as the Japan Federation of Printing Industries, we express our absolute opposition.” As we mentioned in chapter 2, Mitsubishi Corporation bought 30% of the newly issued shares. Hokuetsu Paper does not belong to a horizontal keiretsu and Mitsubishi Corporation does not belong to its largest shareholders. We think that the reason for the purchase of the shares lies in the good relations Hokuetsu has with other companies in the Japan Federation of Printing Industries. Mitsubishi Paper belongs to the Mitsubishi UFJ Financial Group keiretsu of which Mitsubishi Corporation is a member. Mitsubishi Paper, the fifth largest paper company, had a comprehensive Business Alliance with Hokuetsu Paper during the period July 2000 to July 2005. This business alliance might have led to

139

151 Chapter 7 Institutional approach to hostile takeovers

particularistic trust and the above mentioned rescue. The other party that bought new shares of Hokuetsu (8.8% ownership) was Nippon Paper. The latter two companies signed a strategic alliance in which management autonomy was respected on December 1, 2006. When looking at the bidder companies, the hostile tender offers attempts were undertaken by two foreign bleu-chip companies, three Japanese companies, of which one a bleu-chip company, a U.S. investment fund and a Japanese investment fund. Only the hostile tender offers initiated by the blue-chip foreign companies were successful. The main reason for their success seems to lie in the fact that both companies had a long business relationship with the target company. Cable & Wireless was one of the founding companies and largest shareholders of International Digital Communications. Boehringer Ingelberg was the largest shareholder of SS Pharmaceuticals Company and even held a seat on the board of directors. Both companies wanted to strengthen their relationship with the target company as respectively the telecom and pharmaceutical sectors were being deregulated. They wanted to increase their shareholdings in a friendly way with the management but failed and then turned to the hostile tender offer. The interesting aspect in this sequence of events is that it was not perceived as such by the target companies or other related companies. The hostile tender offers succeeded because the companies had shown a strong interest in the target company for a long time and wanted to improve efficiency. Other Japanese companies and also the Japanese government thought it would be beneficial for the Japanese economy, and the hostile tender offers therefore succeeded. This appears to be the largest difference with the other hostile tender offers attempts that all failed. It is unclear how the target companies would be affected if the hostile bidder was successful, and whether it would have a positive influence on the economy. This argument applies especially for the two investment funds that were responsible for eight of the hostile tender offer attempts. It is unclear whether the companies’ intention is greenmail, asset-stripping, or restructuring. The three Japanese companies that initiated hostile tender offers attempts also failed. Of these three companies, Yumeshin Holdings and Don Quijote do not belong to a trade association, and this could be a reason why they initiated a hostile tender offer. The most interesting aspects of these hostile tender offer attempts is that in both cases the hostile bidder did not engage in a hostile tender offer contest with the white knight. Hereupon we can conclude that the hostile bidder companies either did not have a real intention of getting operational control over the target company, or there were other factors that pressured the company from not engaging in a tender offer contest. The hostile bidder companies were able to profit from the sale of the target company’s shares to the white knight, and this can indicate that the hostile takeover was merely a greenmail attempt, similar to the shite groups. In any event, we do not think that the nature of hostile takeovers in Japan is changing as the recent hostile tender offer attempts can (also) be interpreted as greenmail. This, combined with the fact the attempts remain at a very low level, leads us to question the generally accepted explanations and the influence of the unwinding of cross-shareholdings and changes in Japanese cultural characteristics. With our model we therefore propose the vertical keiretsu and trade association as explanations why Japanese blue-chip companies do not initiate hostile tender offers.

140

152 Section 7-4 Concluding remarks

7-3-2 Generated Hypotheses In section 7-2 we explained that previous research argues that there were no hostile tender offers in the post-war period as a result of horizontal keiretsu and the specific Japanese cultural characteristics. In chapter 5 we showed that as of the 1980s the stable and cross- shareholdings between companies reduced in importance. We also explained in chapter 1 that the cultural characteristics such as life-time employment and seniority-based wages are not pursued by Japanese companies any longer. Although these two institutional elements are said to have changed in Japan, we indicated above that we do not see an increase in hostile tender offers. This raises the question whether they are indeed changing and/or have the influence, as claimed in previous research. With our institutional analysis we find that other explanations for the low number of hostile tender offers could be the vertical keiretsu or trade associations. With our inductive research approach we generate the following hypotheses with reference to hostile tender offers in Japan.

Hypothesis I: The horizontal keiretsu and the specific Japanese cultural characteristics did not influence hostile takeovers. Therefore, the observed recent changes do not have any effect.

Hypothesis II: Horizontal keiretsu and/or the specific Japanese cultural characteristics did not change to the degree it influenced hostile takeovers.

Hypothesis III: Hostile takeovers are prevented by vertical keiretsu and/or trade associations.

Hypothesis IV: Combination of Hypothesis II and Hypothesis III.

Hypothesis V: Hostile takeovers are prevented by other factors.

7-4 Concluding remarks In this chapter we looked into interpretations on corporate groups and hostile takeovers from previous research, and we provided our own interpretation based on our institutional model. Previous research argued that the six horizontal keiretsu were formed as a result of hostile takeovers in the 1950s, and contends that these subsequently have prevented hostile takeovers in the entire post-war period. Another explanation put forward is that hostile takeovers do not fit in with the Japanese culture. Regarding the existence of the vertical keiretsu, it is claimed that these can be sustained without contractual safeguards because of a high level of trust, importance of reputation, and financial hostages. Before we provided an explanation for the corporate groups according to our institutional model, we indicated that the Japanese society is characterized by high particularistic trust and low general trust. It is important to understand this trust is very different from for example the U.S., as it results in the corporate groups in Japan. The horizontal keiretsu finds its origins in the ie system and developed from the pre-war zaibatsu. Resulting from the new formal institutions implemented by SCAP after WWII,

141

153 Chapter 7 Institutional approach to hostile takeovers

the pre-war governance structures re-arranged itself in the horizontal keiretsu. The vertical keiretsu is different from the horizontal keiretsu in that it developed after WWII. These groups are characterized by a relationship in which the parent company and subsidiary are mutually dependent. This mutual dependency and the benefit for the parent company, rather than trust, have the most important influence on this relationship. The first trade association was formed in 1092AD and trade associations have been very important in the economic development of Japan. This also applies to companies in the same business sectors, as the trade associations disseminated new technology and information. Trade associations also try to influence government policy and the (competing) member companies developed good relationships with each other. We showed that the horizontal keiretsu as the most important prevention measure against hostile takeovers is doubtful. Based on our analysis we concluded that in the post- war period, all hostile takeovers need to be interpreted as greenmail attempts. New laws implemented after WWII created the possibility to engage in greenmail and, as companies could not repurchase their own shares prior to 2001, companies that were part of a horizontal keiretsu were targeted. Rather than preventing hostile takeovers, the horizontal keiretsu appear to have caused greenmail in the post-war period. Trade associations are formed by companies in the same industrial sector and information is exchanged and relationships are built. We show that because of the strong relations in these groups, companies build a high level of particularistic trust, preventing these companies to engage in hostile takeovers against each other. The vertical keiretsu is another important impediment to hostile takeovers, functioning in three ways. The first characteristic is that a large Japanese company is not the same as a large U.S. company. Whereas the U.S. company tends to produce most products within the company, the Japanese company uses a wide subcontracting system that makes it a difficult target for a hostile takeover. The second characteristic of the vertical keiretsu is that supply from the subsidiaries is indispensable in the production process of the parent company. The majority of its important subsidiaries is already fully owned but if a hostile takeover is attempted on an important partially owned subsidiary, the parent company will intervene to secure its own production process. The third aspect is the question whether it would be beneficial for a hostile bidder to try and takeover an unwilling subsidiary company. Even if the hostile bidder would succeed it is very likely it will lose the subsidiary’s suppliers and customers in the process. All these factors have limited hostile takeover attempts to a high degree.

142

154 8

Introduction to M&A

This chapter introduces M&A structures and discusses merger motives and drivers for merger wave. Section 8-1 explains that M&A can be for the purpose of business diversification and expansion, or corporate restructuring. The section indicates the structures of M&A that can be used for each purpose and explains how a mergers or Tender Offer Bid (“TOB”) can be implemented. Some available defensive measures for hostile takeovers are also described. Section 8-2 discusses four motives for mergers and two merger wave drivers, and presents findings of previous research. An overview of the five merger waves that occurred in the United States since 1890 is given in section 8-3. The discussion of the U.S. merger waves serves as a reference for our own findings on merger waves in Japan. We conclude the chapter with section 8-4.

143

155 Chapter 8 Introduction to M&A

8-1 M&A terminology69 R M&A activity can be aimed at (i) diversification or expansion or (ii) restructuring. R Mergers and tender offers are diversification and expansion activities in which managerial and ownership rights are transferred. R Defensive measures for hostile tender offers are the poison pill, white knight, white squire and greenmail.

M&A generally refers to a broad range of M&A structures including capital participations, business tie-ups, divestures and other forms of restructuring. In sub-section 8-1-1 we provide an overview of the main M&A types. This thesis primarily looks into (i) mergers and (ii) Tender Offers (also referred to as “TOB”, “Tender Offer Buyout/Bid” and “Takeover Bid”). Sub-section 8-1-2 explains the three most important merger structures and sub-section 8-1-3 introduces the tender offer and some defensive measures against hostile takeovers.

8-1-1 Types of M&A Figure 8-1 shows that M&A activity can be divided in two activity types: (i) diversification or expansion and (ii) restructuring. Diversification or expansion activities can be structured as a merger, acquisition, capital tie-up or contractual agreement. Restructuring activities consist of selling a company’s business unit(s) or (a part of all) assets.

Diversification or expansion – Within this type we can distinguish transactions with, and without, a transfer of managerial and ownership rights. Transactions with a transfer of managerial and ownership rights are the merger and acquisition. The merger is discussed in the next sub-section. Here we discuss the two forms of acquisition. In a stock acquisition the acquiring company buys more than 50% of a target company’s shares and herewith attains majority ownership. This can be achieved by (i) stock acquisition from existing shareholders, (ii) subscription of new issues of stock, (iii) acquisition via stock swap or (iv) acquisition by Leveraged Buy-out (LBO), Management Buy-out (MBO) or TOB. An alternative to the stock acquisition is the asset acquisition, in which the buyer company purchases the target company’s assets. The buying company can limit its acquisitions to those parts of the target company that coincide with the buyer’s needs. If the buyer company purchases all the assets of the target company, the latter becomes a corporate shell with cash or securities that it received from the acquisition as assets. After the transaction the company can be dissolved and a liquidating dividend paid to the stockholders. Alternatively, the company may use its liquid assets to purchase other assets or another company. After a partial acquisition of the target’s assets, the target company can be continued after the transaction. Activities not involving transfer of managerial and ownership rights are the capital tie- up and contractual relationship. The first type of the capital tie-up is a capital participation in which the acquiring company does not acquire more than 50% of the outstanding shares. The second type of capital tie-up is a joint venture company, a separate entity in which two (or more) companies invest. The third type is the alliance, but this does not necessarily

69 This section is based on Ravenscraft and Sherer (1987); Mueller (2003); Bruner (2004).

144

156 Section 8-1 M&A terminology

involve equity participation. It can also be an alliance to exchange only human capital or resources. On the other hand, it is more than a contractual relationship. A contractual relationship is a licensing agreement, co-marketing agreement or joint-purchasing agreement.

M&A activity

Diversification or Expansion Restructuring

Merger Sell-off

Statutory merger Spin-off

Statutory consolidation Split-up

Triangular merger Split-off

Acquisition Equity carve-out

Stock acquisition Divesture

Asset acquisition

Capital tie-up

Capital participation

Joint venture company

Contractual relationship Figure 8-1 M&A types

Restructuring - The restructuring type of M&A consists of transactions in which a new entity is created. New entities are created through a sell-off, split-off and equity carve-out. A sell-off is the sale of a company’s business or (a part of all) assets, and can be executed as a spin-off or split-up. A spin-off is a sell-off of a part of the business or assets, through which a separate new legal entity is created. The shares of the new entity are distributed on a pro rata basis to existing shareholders of the original company. The shareholders of the original company will have the same proportion of ownership in the new entity. The main reason of this kind of sell-off is to create a separation in control and decision making between the original company and the new entity. A split-up is a series of spin-offs through which the original company is replaced by newly established entities. The original company ceases to exist after the split-up. A split-off is a restructuring form in which a subsidiary becomes a separate freestanding entity. The parent company of the subsidiary swaps its own shares for shares in the subsidiary. A proportion of the parent company’s shareholders receives ownership of the new entity (previous subsidiary) in exchange for their ownership of the parent company. In an equity carve-out a business unit of a company is sold through an initial public offering (IPO). The business unit forms a new separate entity and its shares are sold to outsiders. The equity holders need not be the same as the equity holders in the original seller. A new control group is immediately created.

145

157 Chapter 8 Introduction to M&A

In a divesture a part of the original company is sold to an outside third party. The original company receives cash or equivalent consideration for the sale. Usually the buyer is an existing company and no new legal entity results. It can be seen as a form of expansion on the part of the buying company.

8-1-2 Mergers Prior to our discussion of the merger types, we look briefly into the standard merger process.

Merger process - During a friendly merger the merging companies’ management, shareholders, board of directors and creditors are involved in the process. Their relative influence primarily depends on the company’s financing decision70 i.e. the management’s choice between internal, debt and equity financing. The financing decision results in a company’s capital structure and has a determining effect on the ownership and corporate governance structure.71 Generally, the bidder company’s management will approach the management of the target company to suggest a merger between the two companies. If the target company’s management is positive towards the proposition, the management of the two companies will enter the negotiation phase. When the two parties reach an agreement about the merger conditions, the merger process generally develops as follows: ï Board of Directors’ meeting; to obtain board approval of the merger proposal and execute the merger agreement. ï Shareholders’ meeting; to obtain approval of the merger agreement. ï Creditors are permitted to object to the merger after the shareholders’ meeting. If the creditors make an objection to the merger, the company must provide them with collateral security, property or debt. ï Registration of changes in corporate structure. The merger process shows involvement of management, shareholders, board of directors and creditors. The latter actors can play an important role in the merger process, especially when companies fall in financial distress. Creditors provide the company debt financing as loans or corporate bonds in the form of a tradable public issue or custom tailored private placement. Debt creates a tax shield for interest and shield benefits increase with a higher amount of debt. The higher level of debt, on the other hand, also raises the probability of financial distress and consequently bankruptcy costs (Brealey and Myers 1996). A company having difficulty to meet its financial obligations is in financial distress. When the company cannot meet its obligations, this can lead to its bankruptcy. In case of bankruptcy, the creditors will take control over the company. Warner (1977) explains there

70 Myers (1984) indicates that according to the pecking order of financing, managers prefer internal finance and, if external finance is required, debt over equity financing. Myers gives the interpretation of managerial capitalism on the preference for internal financing as “a byproduct of the separation of ownership and control: professional managers avoid relying on external finance because it would subject them to the discipline of the capital market” (p.582). The preference of debt over equity results from asymmetric information between the management with (insider) information and the shareholders/investors. A management that is optimistic about the future of the company will consider its equity as undervalued and therefore issue debt. If the management is pessimistic about the company’s future, it will issue the overvalued equity. On the other hand, as the investors will interpret an equity issue as a signal that equity is overvalued, the company’s stock price might fall. The management will therefore decide to issue debt. 71 The influence of the financing decision on the ownership structure is through private placement of shares.

146

158 Section 8-1 M&A terminology

are direct and indirect bankruptcy costs. Direct bankruptcy costs are the legal and administrative costs at the time of the bankruptcy. “[I]ndirect costs include lost sales, lost profits, and possibly the inability of the firm to obtain credit or to issue securities except under especially onerous terms” (p.338). These bankruptcy costs represent the difficulties to run a company that is going bankrupt. It can become impossible to get financing for necessary investments and/or (other) creditors could block selling off unnecessary assets. Creditors can also take control of the company and engage in restructuring and M&A activities.

Statutory merger - In a statutory merger the bidder company absorbs the target company. The target company seizes to exist and its assets and liabilities are taken over by the bidder company. A statutory merger is normally applied when the bidder is significantly larger than the target. The target shareholders receive the bidder company’s shares and/or “boot” (cash or notes) in exchange for their shares. A merger paid with cash will have no influence on the shareholder structure of the bidder company. A payment in shares however, depending on the target company’s shareholder structure, might result in large changes in the voting control of the new company. By utilizing a stock swap, the bidder company can exchange the shares of the target company’s shareholders for its own existing or newly issued shares. A very important feature of the stock swap is that, upon approval of two-third of both companies’ shareholders 72 , the bidder company can compel minority shareholders of the target company to exchange their shares. Also, in transactions where shares are swapped or transferred, there is no need to raise funds for stock. Prior to the legal amendments in Japan during the late 1990s, it was only possible for the bidder company to squeeze out dissenting minority shareholders with cash. If the minority shareholders do not wish to become shareholders of the acquiring company, they have the right to sell the stock of the target company to the acquiring company. Thus, it makes it easier for companies to gain 100%-ownership in other companies for either restructuring or acquisition purposes. It also enables a parent company to easily get 100%-ownership in partially-owned subsidiaries and create a holding company structure (Lebrun 2001).

Statutory consolidation - A statutory consolidation is a combination of two or more companies in which the merged companies seize to exist and an entirely new entity is formed. The shareholders of the merging companies receive shares of the new company in exchange for their shares, which is also referred to as ‘stock transfer’. A consolidation is often used in case of a “merger of equals”. In the literature and practice, this combination is also frequently referred to as merger.

Triangular merger - In a triangular merger the bidder company establishes or has a subsidiary company to merge with the target company. The bidder company can provide the subsidiary with cash or its own shares as a medium of payment for the merger. A forward triangular cash merger is a merger in which the subsidiary company buys the shares of the target company with cash and then absorbs the target. The target company seizes to exist. In a reverse triangular cash merger the payment is the same but in this case

72 ᨓith at least 50% of all outstanding shares present at the shareholder’s meeting.

147

159 Chapter 8 Introduction to M&A

the subsidiary company is merged into the target company. The subsidiary company seizes to exist. The other possibility is that the subsidiary uses the bidder company’s shares as a medium of payment. This structure has two advantages for the bidder company compared to a statutory merger. Firstly, the target company’s liabilities will be transferred to the subsidiary, leaving the bidder company insulated from risks. Secondly, the merger does not need to be approved by the bidder company’s shareholders. In a forward triangular merger the subsidiary company buys the shares of the target company with the bidder company’s shares and boot. The target company is absorbed by the subsidiary. In a reverse triangular merger the subsidiary company buys the shares of the target company with the bidder company’s shares and boot. Subsequently, the subsidiary company is merged into the target company.

8-1-3 Tender offers In a tender offer the bidder company attempts to gain a controlling interest in the target company by asking the target’s stockholders to submit (tender) their shares. The tender offer is an offer outside the public market, often against market price at a premium, and can be made to the target’s board of directors or directly to the shareholders. When a bidder company expects the management and board of directors to be neutral or positive to a tender offer, it can consider a bear hug approach. According to this approach, the bidder will make the tender offer to the target’s board of directors without making a public announcement. The board of directors will be informed of the bid proposal and be required to make a quick decision. If the target company’s board of directors and management support the bid it will be a friendly tender offer, if they do not support the bid it will be a hostile tender offer. In the latter case the bidder company will ask the target’s shareholders to tender their shares without the board of directors’ approval. If the bidder could have expected the negative response, it could also have decided to approach the shareholders immediately, not engaging in the bear hug. Schwert (2000) indicates that “most transactions contain elements of both friendly and hostile deals. That is, some stakeholders are likely to be disadvantaged by the transaction and there are likely to be some economic gains from combining the operations of the bidder and target” (pp.2599-2600), and that “[H]ostility is usually perceived when an offer is made public that is aggressively rejected by the target firm” (p.2600).

8-1-3-1 Defensive measures There are numerous defensive measures the target company can take in order to prevent a hostile takeover from succeeding.

Poison pill - The target company issues rights to existing shareholders to obtain new shares at nominal cost (also referred to as ‘rights plan’). When a bidder company acquires more than the threshold percentage of shares under the rights plan, the rights allow all shareholders (excluding the bidder who triggered the threshold) to convert the right into common shares. This dilutes the bidder company’s ownership percentage in the target, and makes it more expensive and difficult to acquire control of the target company.

White knight - When a hostile tender offer is being carried out, the target company may try to find a white knight. This is a friendly bidder company that will merge with the target

148

160 Section 8-2 Merger motives

company agreeing not to break it up nor drastically cut the labour force. The white knight either has a business relationship with the company or is a direct competitor of the target company. The white knight could be motivated by his own strategic motives, or just to prevent the hostile bidder taking over the target company.

White squire - A friendly acquirer purchases a large block of stock in the target. As a large stable shareholder, the white squire agrees to vote supportive for the target company’s management but does not take control of the target company.

Greenmail - An acquirer (green mailer) buys up shares of a target company with the objective that the target company will repurchase them at a premium. In these transactions the green mailer makes profits at the expense of the company’s shareholders.

8-2 Merger motives R The synergy motive implies that synergy gains are realized when two companies are combined. R A merger with an adaptive (failing company) motive can be seen as an alternative to bankruptcy. R Mergers with a managerial empire motive are driven by the management’s personal goal at the expense of the company’s shareholders. R The hubris motive implies that a bidder company’s management overestimates obtainable synergies and therefore overpays.

A merger motive applies to individual merger cases and explains why a company decides to merge with another company or to take over a company. In this section we discuss the synergy, adaptive (failing company), managerial empire building, and hubris motive. If applicable, we also examine the influence of the type of merger. Three merger types can be distinguished; (i) the horizontal merger is a combination of companies in the same line of business, (ii) the vertical merger is a merger between companies in different stages of the same production cycle, and (iii) the conglomerate merger is a combination of companies that are engaged in unrelated types of business.

Synergy motive - The synergy motive implies that synergy gains are realized when two companies are combined. The combination of the two companies results in a higher value than the value of the companies on a stand-alone basis (Jensen and Ruback 1983; Healey, Palepu and Ruback 1992). This motive is most important in the discussion on mergers and acquisitions and is a rational motive, based on expectations of possible gains. According to this motive, the management of bidder and target companies will only engage in M&A activity if it maximizes shareholder wealth of both companies. The wealth gains to the target company’s shareholders should therefore be positively correlated to the bidder company’s shareholder wealth gains, and both these gains will be positively correlated with total gain (Berkovitch and Narayanan 1993). Berkovitch and Narayanan (1993), Bradley et al. (1988), and Goergen and Renneboog (2003) find support for the synergy motive in transactions with a positive total gain. Synergy can result from increased market power or efficiency. Horizontal mergers can increase market power of the bidder company by reducing the number of competitors. It is

149

161 Chapter 8 Introduction to M&A

also possible that costs are reduced through higher efficiency by economies of scale and/or stronger bargaining power towards suppliers, customers or financiers. On the other hand, this motive can get a lot of opposition or even be prohibited by the government when it has a large influence on the concentration within an industry. A vertical merger can achieve higher market power by making it more difficult for other companies to enter the vertical production chain, and by securing the company’s supply and/or sales (Comanor 1967). Merging with a supplier is called backward integration, and a merger with a company at the sales side is called forward integration. In a vertical merger better efficiency is possible by (i) elimination of transaction costs between the two companies (no more profit seeking by either company) (Williamson 1975), and (ii) by a shortened production process after the merger. In a conglomerate merger market power can be increased by diversifying the company’s activities through mergers. The company’s overall performance will be less dependent on developments in one particular business sector (Mueller 2003; Bruner 2004), and companies can increase efficiency with economies of scope (Bruner 2004; Ravenscraft and Sherer 1987).

Adaptive (failing company) motive - In some cases a merger is a good alternative for bankruptcy. Dewey (1961) argues that most mergers “have virtually nothing to do with either the creation of market power or realization of scale economies. They are merely a civilized alternative to bankruptcy or the voluntary liquidation that transfers assets from falling to rising firms” (p.257). In this case the creditors will play an important role in the restructuring activities to secure their outstanding loans. The generalization of the adaptive motive by Dewey (1961) is not supported by research that examines the profitability of target companies. Mueller (1980) and Harris et al. (1982) find that target companies have average profit rates similar to non-target companies and bidder companies. Weston and Mansinghka (1971) and Melicher and Rush (1974) investigate conglomerate mergers during the 1960s in the U.S., and show that bidder companies had below average profits, which were also lower than those of target companies. Ravenscraft and Scherer (1987, Ch.3) find that their sample of target companies had significantly higher profit rates than similar non-target companies. Clark and Ofek (1994) look into 38 takeovers involving a distressed target and do not find that the acquisition leads to superior performance. On the other hand, they cannot conclude that the restructuring of the target company was a poor choice in relation to other courses of action. Lang, Stulz, and Walking (1989) and Servaes (1991) examine returns in mergers and takeovers involving poorly performing targets with Tobin’s q ratio as a measure of managerial performance. Servaes (1991) finds that target, bidder and total takeover returns are larger when the bidder company is performing well and the target company is performing poorly. Holl and Kyriazis (1997) investigate UK mergers and their findings support Servaes (1991) in that well-managed bidder companies acquire under- performing target companies to obtain wealth gains.

Managerial empire building - The managerial empire building theory refers to the behaviour of managers regarding M&A transactions. When management of a bidder company decides to merge with another company, it is possible that this is driven by the management’s personal goal at the expense of the company’s shareholders. The managerial empire building motive argues that managers knowingly overpay in M&A deals that are motivated by a maximization of the managers’ own utility at the expense of

150

162 Section 8-2 Merger motives

the company’s shareholders. The self-interest of the bidder’s management is the prime reason for an M&A deal. The management might decide on a merger or acquisitions because of the pursuit of growth in physical size of the company. The increase in size of the company can result in more power and prestige for the manager and also have a positive effect on his remuneration. The motivation of the merger is not the creation of shareholder wealth, and the growth maximization hypothesis predicts that the abnormal return will be low or negative when this policy becomes known to the market (Goergen and Renneboog 2003; Conyon and Murphy 2002; Mueller 1969; Mandelker 1974; Malatesta 1983). The managerial empire building motive is reflected in the bootstrap game and use of surplus funds for M&A. Bootstrap game - The bidder company can decide to merge with a target company to increase its share price based on a higher accounting profit. This is the so-called bootstrap game in which a bidder company with a high Price/Earnings-ratio merges with a target company that has a low Price/Earnings-ratio. If the bidder company pays in stock, it will increase its earnings per share and this can result in a higher share price if the market is fooled (Brealey and Myers 1996). Surplus funds - Companies that generate a substantial amount of cash but do not have any profitable internal investment opportunities, look for ways to spend this capital. Ideally such a company should distribute the surplus funds to shareholders by increasing its dividend payments, or it should repurchase its stock. However, many companies turn to mergers or acquisitions (Bruner 2002). This is not efficiency-related, but is used to prevent the capital from flowing out of the company (Jensen 1986).

Hubris - The hubris theory also refers to behavioural assumptions and argues that the bidder company’s management overestimates the obtainable synergies and therefore overpays the M&A deal. Hubris, excessive pride and arrogance, refers to the assumption of the management of the bidder company that it can create value where others cannot (Roll 1986). Total wealth of the deal is zero and wealth is transferred from the bidder company’s shareholders to the target company’s shareholders. Berkovitch and Narayanan (1993) argue that an M&A deal motivated by hubris will have a negative correlation between target and bidder gains, and zero correlation between target and total gains. Berkovitch and Narayanan (1993), Roll (1986), Malmendier and Tate (2003), and Hayward and Hambrick (1997) find evidence for the hubris amongst bidder companies in the U.S. Raj and Forsyth (2003) examine acquisitions in the UK market and find that bidder companies pay a higher premium in hubris acquisitions than general acquisitions.

151

163 Chapter 8 Introduction to M&A

8-3 Merger waves R Merger waves can be driven by (i) external shocks, such as shocks to an industry’s economic, technological or regulatory environment or (ii) overvaluation of stocks. R Five merger waves occurred in the U.S.; the waves were driven by a combination of factors but ended when the stock market crashed. R The merger motives differed in each wave, which was primarily caused by laws and regulations related to market concentration and financial markets.

Merger waves are periods when, relative to other periods, a lot of M&A activity occurs. In this section we discuss two drivers of merger waves and the merger waves that occurred in the United States.

8-3-1 Drivers for merger waves A driver for merger waves is different from a merger motive as it explains an overall M&A movement, rather than an individual merger case. Merger waves can be driven by external shocks or overvaluation in stocks. Although these are interrelated, each approach places its emphasis on a particular aspect.

External Shocks - The neoclassical theory predicts that merger waves are initiated by shocks to an industry’s economic, technological or regulatory environment. After an industrial shock has taken place, assets will be reallocated as quickly and efficiently as possible in the form of M&A activity (Andrade and Stafford 2004; Mitchell and Mulherin 1996). Mitchell and Mulherin (1996) document that deregulation, increased foreign competition, financial innovations and oil price shocks can explain takeovers and restructurings in industrial sectors. Andrade and Stafford (2004) make a distinction between the “expansionary” and “contractionary” role of mergers; the former increases capital stock of a company, similar to an internal investment, the latter facilitates consolidation and reduction of a company’s asset base. According to the expansion hypothesis, mergers can be seen as a way to respond to increasing economic growth and a positive business cycle. The merger can increase the company’s market power or efficiency and provide possibilities to exploit the market. “[T]he profits of such market control are of course all the greater if the market is expanding. We might thus expect attempts at mergers for market control to occur early in a cyclical expansion, when expectations become favorable” (Nelson 1959, p.107). According to the contraction (retardation) hypothesis, on the other hand, mergers are a means by which a company can preserve profits when economic growth is falling. Mergers occur when the overall business cycle is negative, demand falls or competition is rising. Jensen (1993) argues that it can be the result of changes in the product and factor markets, legal, political and regulatory systems, and the capital markets. In particular, he indicates that excess productive capacity in many industries caused by technological and supply shocks will result in reorganization and restructuring activities in order to eliminate the excess capacity. Harford (2005) argues that a merger wave can only result from an external shock if there is sufficient capital liquidity to accommodate reallocation of assets.

152

164 Section 8-3 Merger waves

Overvaluation of stocks - The behavioural theory takes market mispricing as a given and argues that companies’ market valuation is most important for M&A activity. According to this theory managers are rational and the stock market is irrational. Managers will engage in M&A activity when their own company is overvalued. The overvalued stock effectively lowers the price of the acquisition and is used to acquire the assets of undervalued or less overvalued companies. According to this theory, merger waves occur during stock market booms as a result of managerial timing of market overvaluations of their companies (Shleifer and Vishny 2003). Rhodes-Kropf and Vishwanathan (2004) build a theoretical model that assumes that managers of bidder companies have complete information on the value of the company and synergies that will be created with a merger. The managers of the target company, on the other hand, do not have this information. In a stock market boom, bidder companies will use stock swaps to finance the M&A activity with overvalued stock. The managers of the target company will accept the bidder company’s offer financed by the overvalued stock because they do not have complete information.

8-3-2 Merger waves in the United States In the United States and Europe five merger waves can be distinguished since 1890. In this section we look into the waves that occurred in the United States and explain their most important characteristics.73

1,400 2,500 First wave Second wave

1,200 2,000

1,000 muti l.dollar mln. in Amount

1,500 800

600 Merger cases 1,000 s 400

500 200

0 0

5 7 3 9 5 7 1 3 9 89 89 90 90 911 91 91 92 92 92 1 1 1899 1901 1 1905 1907 1 1 1913 1 1 1919 1 1 1925 1927 1 M&A announcements Transaction amount

Figure 8-2 Mergers in the United States in the period 1895-1930 Note The graph depicts the merger cases that occurred in the manufacturing and mining sectors. Source Nelson (1959) for 1895-1918. Eis (1969) for 1919-1930

First merger wave - The first merger wave in United States started in 1895 and ended in 1904. The average annual number of companies that were absorbed into mergers during

73 Please refer to Goergen and Renneboog (2003) for a discussion of merger waves in Europe.

153

165 Chapter 8 Introduction to M&A

this decade was 301 with an average capitalization of 691 million dollars. Merger activity peaked in the years 1899 and 1901 with capitalizations exceeding 2 billion dollars. The companies’ main M&A motive during this wave was to increase market power. The mergers were predominantly horizontal and had a very important impact on the concentration in a wide variety of industrial sectors. Nelson (1959) explains that “[I]t transformed many industries, formerly characterized by many small and medium-sized firms, into those in which one or a few large enterprises occupied leading positions” (p.5). In this period industrial giants like DuPont Inc., Standard Oil, General Electric, Eastman Kodak, and American Tobacco Inc. were established. Gaughan (1999) explains that “[W]hile these companies are major corporations today with large market shares, some were truly dominant firms by the end of the first merger wave” (p.22). The wave is therefore also termed as “merging to form monopolies” (Stigler 1950). Stigler (1950) argues that legislation on incorporation and elimination of almost every restriction on mergers, together with the development of trading in industrial stocks on the New York Stock exchange, had an important influence on the first merger movement. This wave came to an end around 1903-05, when the equity market crashed.

14,000 1,600,000 Third Wave Fourth Wave Fifth Wave

12,000 1,400,000

1,200,000

10,000 dollar million in Amount

1,000,000 8,000

800,000 6,000 600,000 M&A announcements 4,000 s 400,000

2,000 200,000

0 0

3 5 1 3 9 1 7 9 5 7 5 96 96 97 97 97 98 983 98 98 991 99 99 999 00 1 1 1967 1969 1 1 1975 1977 1 1 1 1985 1 1 1 1993 1 1 1 2001 2003 2 M&A announcements Transaction amount

Figure 8-3 M&A announcements in the United States in the period 1963-1998 Source Mergerstat

Second merger wave - The second merger wave started in the boom years after WWI in 1919, peaked in the period 1928-29, and ended in 1930. The average annual number of companies that disappeared into mergers during this twelve-year period was 250, with an average capitalization of 1,012 million dollars. Stigler (1950) explains that the companies that undertook the mergers in the second wave were not the dominant companies in the industry like the first wave. The stricter antitrust environment resulted in mergers of second rank companies. Vertical mergers were

154

166 Section 8-3 Merger waves

most important and industries were “changed from near-monopoly to oligopoly” (p.31). Gaughan (1999) argues that this development was also caused by the “federal government’s encouragement of the formation of business cooperatives to enhance the nation’s productivity as part of the war effort” (p.29). Debt had become an important medium of payment for mergers during this wave. “Black Thursday”, the market crash on 29 October 1929, put an end to the wave.

Third merger wave - The third merger wave started in 1965 and ended in 1970. Gaughan (1999) reports that the average annual number of companies that disappeared into mergers during this six-year period was 3,866 with a peak in the year 1969 when 6,107 mergers took place. The targets were significantly smaller than the bidders in this period. In the 1960s the government’s fear for companies attaining monopoly power was high and tough antitrust laws were implemented. Horizontal and vertical mergers were virtually impossible for larger companies and those that wanted to expand operations therefore turned to conglomerate mergers. M&A activities during this wave were focused on diversification and creation of large conglomerates. The Dow Jones Industrial average had increased from 618 to 906 in the period 1960-68 and made stock an important medium of financing for M&A.

Fourth Merger Wave - The fourth merger wave started in 1981 and ended in 1989. The average annual number of companies that disappeared into mergers during this nine-year period was 2,534 with a total dollar value paid of 147 billion dollars. In this period the average merger was worth 58 million dollars, marking a tremendous increase in value compared with the average of 3.2 million dollars in 1970 (the year after the third wave).

As these numbers indicate, the fourth wave is characterized by mega-mergers (mergers between very large companies). The other characteristic is the higher frequency of hostile takeovers as shown in table 8-1. In the U.S., hostile takeovers already occur since 1908, but during the 1970s major established companies started engaging in hostile takeovers (for example, INCO versus ESB, and United Technologies versus Otis Elevator).

155

167 Chapter 8 Introduction to M&A

Table 8-1 Tender offers in the United States in the period 1980-1993 Year TOB Contested TOB 1980 53 12 1981 75 28 1982 68 29 1983 37 11 1984 79 18 1985 84 32 1986 150 40 1987 116 31 1988 217 46 1989 132 28 1990 56 8 1991 20 2 1992 18 2 1993 32 3 Source Mergerstat

The corporate raider became more prominent during this wave as well. Development of new financial instruments and markets (e.g. the junk bond market) facilitated the financing of mergers and acquisitions. The most important medium of payment were cash and debt and the new debt securities allowed smaller companies to acquire relatively larger targets through leveraged buyouts.

Fifth Merger Wave – The fifth wave started in 1992 and is characterized by deals aimed at strategic motives with bidder and target companies operating in the same industry. Figure 8-3 shows an upward trend in M&A cases up to 2000. The number of M&A deals increased from 2,574 cases in 1992 to 11,123 cases in 2000, the amount increasing from respectively 96.7 to 1,268.6 billion dollars. This is a considerable increase compared to the fourth wave in which the annual average was 2,534 deals at a dollar value of 147 billion dollars. M&A deals showed a sharp drop in the years 2001-2002 as stock prices plummeted on exchanges in the United States, resulting from the September 11 attacks, the bursting of the internet bubble and the accounting scandals. M&A activity has been increasing since 2003 and reached its highest level in 2006 with 11,750 M&A cases at a value of 1,484.3 billion dollars.

8-4 Conclusion This chapter introduced the M&A terminology that is used in the thesis. M&A activities can be divided in the activities related to (i) diversification or expansion and (ii) restructuring. The main focus of this thesis is M&A activity structured as mergers and hostile takeovers. Regarding hostile takeovers we showed some available defensive measures such as the poison pill, white knight, white squire and greenmail. Especially the latter two defensive measures have played an important role in post-war Japan. Recently the poison pill has become popular. We discussed some merger motives and distinguished synergy, adaptive, managerial empire building, and hubris. The synergy motive implies that synergy gains are realized when two companies are combined, whereas a merger with an adaptive (failing company)

156

168 Section 8-4 Conclusion

motive can be seen as an alternative to bankruptcy. Mergers with a managerial empire motive are driven by the management’s personal goal at the expense of the company’s shareholders, and the hubris motive implies that the management overestimates obtainable synergies and therefore overpays. Section 8-3 showed that merger waves can be driven by external shocks, such as shocks to an industry’s economic, technological or regulatory environment, or overvaluation of stocks. Five merger waves occurred in the U.S., of which the first wave (1900) was aimed at increasing market power, and the second wave focused on restructuring industrial sectors. The third wave in the 1960s was aimed at diversification and creation of large conglomerates, whereas the fourth wave was dominated by mega mergers (between large companies) and by hostile takeovers. The fifth wave is characterized by deals initiated for strategic motives.

157

169 158

170 9

Merger activity

This chapter examines the influence of some institutional elements on merger activity in Japan. In section 9-1 we start with an interpretation of M&A activity in the pre-war and post-war periods; we focus on the effect of Japan’s political and economic development, and the role of actors. We look into merger motives and drivers for merger waves, and for the post-war period we discuss some results from studies that look into efficiency effects of Japanese mergers. Section 9-2 discusses previous research on the monitoring role of the main bank, the abnormal return in Japanese and U.S. mergers, and the profitability drivers of mergers. Sections 9-3 and 9-4 show the results of our own studies on shareholder wealth. We examine the influence of Japanese institutional elements such as the main bank system, horizontal keiretsu, and presence of a parent company, and also look into data such as the companies’ profitability and financial distress. In section 9-3 we examine the effect of mergers on shareholder wealth of listed bidder companies in the period 1993-2003. Section 9-4 shows how shareholder wealth of listed bidder and target companies is affected by mergers during the period 1981-2003. Section 9-5 summarizes and concludes the chapter.

159

171 Chapter 9 Merger activity

9-1 M&A activity over time R In the pre-war period no merger waves occurred, and the main motives for mergers appear to have been synergy and adaptive. R The two merger waves in the post-war period were driven by external shocks: the first wave by new regulations and increased competition, and the second wave by a collapse of the stock and land prices (first half 1990s) and new regulations (latter half 1990s). R Merger motives in the post-war period appear to be similar to the pre-war period.

Chapter 2 described M&A activity in the pre-war and post-war periods and provided evidence that pre-war M&A activity was low. Regarding the post-war period we could identify two merger waves: the first wave in the 1960s and the second wave starting in 1991. In this section we revisit our findings and examine merger motives, drivers for merger waves, and which actors played a role. Pettway and Yamada (1986) explain that in the post-war period, mergers between Japanese companies are normally initiated by either a third party or the target company seeking managerial or financial assistance. This third party can be the government, banks, or business leaders and, in contrast to the U.S., the bidder company rarely takes the initiative.

Pre-war - Based on our analysis of M&A activity by 537 listed companies, we did not find merger waves in the pre-war period. Understanding mergers in this period therefore needs to be found in motives for individual mergers. Most M&A activity occurred in the energy sector and we think this can be explained by the pursuit of synergies through economies of scale. The electrification of Japanese companies necessitated improvement of efficiency and M&A activity was indispensable in this process. Also, the 1920s was a period of financial crises and economic recession in which a lot of companies faced difficulties. Most mergers during the 1920s can therefore be explained by the adaptive motive, where weak companies were rescued by stronger companies through M&A. The question regarding the hostility needs to be viewed from this perspective as well. Although it is possible that the difficulties in the 1920s resulted in companies engaging in hostile takeovers of other companies, we think the explanation that companies in financial distress asked assistance from stronger companies or banks is more plausible. This would imply that prior to WWII, mergers were also arranged by a target company or a third party, as argued by Pettway and Yamada with reference to the post-war period. It is often argued that zaibatsu banks were not as important as the post-war main bank because the zaibatsu companies did not rely on loan financing. We think that the analysis of the influence of banks in the pre-war period should not be limited to the financing decision of companies. The post-war main bank system is named as such because the main bank, apart from being a loan provider, has various relations with the company. We therefore argue that the independence on loans should not lead to the conclusion that the main bank was absent in the pre-war period. As the case studies of mergers by Okazaki in chapter 2 showed, banks played an important role in corporate governance; bank executives were directed to become executive director of companies in mergers. Further, the discussion of the importance of the main bank should be directed to the target company, rather than the bidder company. We will come back to this issue in section 9-4 with reference to the second merger wave, but it is important to realize that when a company is in financial distress, its (main) bank will be concerned with its interests and try to arrange a

160

172 Section 9-1 M&A activity over time

merger with a stronger company. Regarding the pre-war period, it is commonly accepted that the zaibatsu bank did not have a lot of its loans outstanding at companies that belonged to its own zaibatsu. If the non-zaibatsu companies to which it lent would fall into financial difficulty, however, the zaibatsu bank is very likely to take the initiative to arrange a merger with a strong zaibatsu company to secure its own interests. Overall, the synergy and adaptive motive appear to have been most important in pre-war mergers. The large shareholders (holding company) played the most important role, but we think that it is very likely that banks also had an important influence on mergers when a target company was in financial distress. This thesis does not allow us to elaborate, but it is very interesting for future research on mergers and the financial system in pre-war Japan.

First merger wave - The first merger wave in Japan occurred in the period 1963-72. The drivers of this wave were diverse, but all were external shocks. The first merger wave coincides with trade liberalization and removal of controls on capital transactions in 1964. These mergers were aimed at preventing excessive competition between domestic companies and increasing their international competitiveness. Acceptance of article 14 of GATT, article 8 of IMF, and becoming a member of the OECD were important pressures for change, but so was the continuing drive in improving production and product quality by companies. It is often argued that the opening of Japan resulted in companies to engage in M&A activity to protect themselves of being taken over by foreign companies, but M&A activity was probably more strongly related to strengthening of competitiveness; this period is also characterized by the formation of the vertical keiretsu, as discussed in section 7-2-3. Further, fluctuations in economic growth in this period resulted in independent decision making by companies, which needs to be included in the analysis. Important actors influencing M&A activity in Japan were business managers, the government and the bureaucracy. Regarding the bureaucracy, especially MITI tried to influence M&A activity through administrative guidance to prevent excessive competition. Evidence on the actual influence of MITI on M&A activity is mixed. Her involvement can be seen in the mergers between Fuji Steel and Yawata Steel, and Prince Motors and Nissan Motors. Friedman (1988) indicates, however, that MITI’s initial plan regarding the automotive industry was to build an industry based on two large companies. The automotive companies rejected these plans and subsequently more automotive companies were established. So, MITI played an important role in post-war M&A activity, but it was certainly not the only important actor in mergers; its influence was highly dependent on acceptance of guidance by the business community. Research on accounting performance 74 of mergers during the first wave primarily indicates that performance did not improve after the merger. Only Ikeda and Doi (1983) find that most of the performance measures improved for the bidder companies after the merger, relative to the matched non-merging companies; they examine 49 mergers of listed companies in the period from 1964 to 1975. In contrast, Muramatsu (1986) looks into accounting performance of 43 mergers in the period 1966-1979 and finds that bidder companies’ profit measures after the merger were lower. He compares these measures with

74 An accounting study examines the effect of a merger based on accounting information. Performance of the mergers is measured as profit rates and growth rates. A comparison is commonly made between the sum of the merging companies before the merger and the newly formed company after the merger. The studies are often organized by comparing the performance with a non-merging matched sample to eliminate the effects of general business conditions and industrial differences.

161

173 Chapter 9 Merger activity

a matching group of non-merging companies, and indicates that the growth rate was not affected. Hoshino (1982) examines 15 mergers in 1970 and reports lower profitability and safety after the merger (net worth to total liabilities and assets, and net profit to total liabilities and assets, turnover ratio, and debt-equity ratio), and higher liquidity (current ratio). He concludes that the motivation for mergers in the sample is managerial empire building rather than pursuit of synergies to create shareholder wealth. It should be noted, however, that he does not find a clear distinction between merging and non-merging companies.

Second merger wave - The second merger wave is driven by external shocks as well. The first period of the second merger wave occurred after the stock market and land price bubbles burst, the economy fell in a recession, and a lot of companies and financial institutions faced financial difficulties. In this period the most important actors were companies that tried to restructure and re-organize their own operations and the operations of subsidiaries and related companies. Banks, facing financial problems themselves, attempted to limit the amount of their non-performing loans and also took the initiative in arranging mergers to secure their own interests. This latter point will become clear in our own empirical study in section 9-4. The merger motives appear to be similar to the motives of the pre-war period, the synergy and adaptive motive being most important. The main actors were managers of companies in financial distress and main banks. The second period of the second merger wave starts after the regulatory changes in the latter half of the 1990s. The number of M&A deals and the total size of M&A activity increase considerably, but the number of mergers remains relatively low and stable in the period 1999-2006. Regarding this wave we look into the role of the main bank and large shareholders with event studies in the sections 9-3 and 9-4. We pay special attention to the role of the main bank that is argued to be important in monitoring of companies in the post-war period. We examine whether the dissolving of cross-shareholdings has indeed lowered the influence of the main bank as is commonly argued. Previous research confirms that the wave is driven by external shocks, but evidence for the merger motive is mixed. Arikawa and Miyajima (2007) investigate the period from 1991 to 2004 and find that the second merger wave in Japan was caused by economic shocks. They demonstrate that during the wave the role of mergers was both reactionary and expansionary; industries with negative shocks (negative changes to growth opportunities and the decreasing sales) experienced larger M&A deals as well as industries with positive shocks. At a company level they find that target companies are companies with lower growth opportunities and high leverage, which suggests that M&A is used as a means of corporate restructuring during this wave. Lin et al. (2008) examine whether bidder companies in Japan are motivated by hubris. Their sample contains of 4582 M&A activities during the period 1989 to 2003, of which 3,005 were with Japanese targets and 1,577 with non-Japanese targets. They find that high (low) hubris bidder companies have negative (positive) abnormal returns and conclude that this is largely consistent with the hubris hypothesis that argues that over-confident managers engage in value-destroying M&A activities. In their investigation of sub-sample periods they find that hubris is more likely to occur during the period 1999-2003 than 1989-1998.

162

174 Section 9-2 Previous research

Kruse et al. (2007) investigate 69 mergers of companies listed on the Tokyo Stock Exchange during 1969 to 1999. Because they do not provide results for sub-periods, we discuss their results here, but they cannot be seen as representative for the second merger wave. They find that operating performance improves after the merger, especially for inter- industry mergers. They only provide information between time-periods (1969-89 versus 1990-99) related to changes in employment and find that these are positively related to post-merger performance for mergers in the period 1969-89. There is no relation between operating performance and changes in employment during 1990-99, existing relationships among merging companies, and distressed targets. Studies that primarily focused on the period prior to the second merger wave find negative performance following the merger. Odagiri and Hase (1989) examine profit rate change and growth rate change in 243 M&A cases during the period 1980-1987. The M&A cases consist of 43 mergers, 135 acquisitions and 65 capital participations. They do not find significant evidence that the M&A cases improved profitability or growth of the bidder companies. Yeh and Hoshino (2002a) examine 86 corporate mergers in the period 1970-1994 with a listed bidder, of which in 30 cases the target company was also listed. They compare various accounting performance measures in pre-merger and post-merger years with industry medians, and find that mergers in general have a negative impact on the performance measures. Examination of the difference between keiretsu related mergers and independent mergers shows that the keiretsu related mergers have a considerably stronger negative influence over profitability and sales growth.

In contrast to the pro-cyclical U.S. merger waves, we find that mergers in Japan tend to be counter-cyclical, both with respect to the general economy as well as with respect to stock market valuations.

9-2 Previous research R The main bank plays an important role in monitoring companies and, as a result of its strong relationship with the company, can intervene quickly when the company faces financial difficulty. R Whereas positive abnormal returns for U.S. target companies at the announcement date are considerable, previous research on Japanese target companies shows predominantly negative results (or only slightly positive).

In this section we discuss previous research on main bank monitoring and abnormal return at merger announcements, and turn to our own empirical research in the next sections.

9-2-1 Main bank monitoring Aoki, Patrick and Sheard (1994) find that companies affiliated with a horizontal keiretsu are more efficiently monitored by the main bank because short-term bank loans reduce information asymmetry. Stulz (1990) and Kang, Shivdasani and Yamada (2000) argue that short term debt and the management of payment settlement accounts allow banks in general, and main banks in particular, to obtain confidential information about firms’ prospects and investment opportunities. It also allows financial institutions to intervene quickly when problems arise. This results in lower bankruptcy costs and the main bank’s presence also reduces free-rider problems between creditors (Berglof and Perotti 1994).

163

175 Chapter 9 Merger activity

Prowse (1990) shows that the threat of equity holders taking risky projects in order to expropriate wealth from debt holders is eliminated in Japan due to large equity positions of financial institutions, and especially of the main bank. Consistent with the main bank story, high debt costs are attributed to an “agency fee” for bank monitoring by Aoki (1988) and to an insurance premium for bailing out by the main bank in case of financial distress by Hoshi, Kashyap and Scharfstein (1990). Nakatani (1984) argues that the economic rent that the main bank extracts is an insurance premium against financial distress of the members. The long-term relationship with group-companies and risk-sharing between the company and the main bank (sticky interest rate) make it very difficult for the bank to default on its obligations if the company falls into difficulties. The threat of competition in the capital market is herewith reduced as well. Keiretsu companies in financial distress are also less affected by capital rationing and illiquidity (Hoshi, Kashyap and Scharfstein 1991). Kaplan and Minton (1994) find that when a company’s stock market performance is poor, but the company is not in financial distress yet, member companies of the horizontal keiretsu and the main bank will appoint external members on the corporate board (strongly related to shareholdings). Once the company is in financial distress, only the main bank will send external directors and take effective control of the company. Japanese banks are herewith a substitute for shareholder pressure. Kang and Shivdasani (1995) also show that new bank representatives are appointed in the boards of Japanese companies when their financial performance lags. Aoki (1990), Kester (1991), and Kaplan (1994) argue that Japanese executives manage companies in the interest of debt-holders, and that, conditional on sufficient earnings to satisfy banks and meet debt payments, managers can run companies in their own and their employees’ interests without interference from the main bank. Jensen (1989) argues that financial institutions holding debt and equity participation in the company, are inclined to monitor company’s management strictly, which creates strong incentives for management to make value-maximizing decisions. Morck and Nakamura (1999) suggest that banks collude with managers of companies to deter external threats to corporate control and to collect rents on bank loans. Horiuchi, Packer and Fukuda (1988) argue that the main bank relationship seems to be less stable than generally believed. Change of main bank is caused by company growth and not by uncertainty of performance. They do not find evidence that the main bank relationship stabilizes performance of borrowing companies as a result of risk-sharing. They mention that the insurance function of the main bank might come into play only when a borrowing company gets into financial difficulties. As not all companies in financial distress are bailed out or saved, however, the authors argue that the main bank makes an assessment of the company’s economic viability at her own discretion. Morck and Nakamura (1999) show that Japanese banks act primarily to protect their interests as creditors, responding to potential and actual debt repayment problems rather than more general indicators of financial health.

9-2-2 Previous event studies Previous research on Japanese domestic mergers shows that bidder companies have a positive stock price effect up to the announcement date of the merger, but this effect turns negative thereafter. Similar results are found for target companies; the abnormal returns before the announcement date are only slightly more positive for bidder companies. Pettway and Yamada (1986) examine the period 1977-84 and find, for their 16 observations, positive returns for the bidder companies around the announcement date. The

164

176 Section 9-2 Previous research

abnormal return at [-1] is significant at 0.60% and the Cumulative Abnormal Return (CAR) for the 2-day period [-1,0] is positive at 0.70%, but insignificant. Ito (1989) investigates 31 mergers between listed companies in the period 1971 to 1987, covering the period of Pettway and Yamada. He finds a significantly positive CAR of 1.15% for the period [-1,0]. Although insignificant, the CAR becomes negative when the period around the announcement date is expanded. This negative abnormal return for the expanded period is confirmed by Komoto (2002) who examines mergers in the period 1980 to 1999. In his sample of 88 mergers between industrial companies, he finds a negative CAR of -2.1% for the period [-5,+5]. Yeh and Hoshino (2002) investigate 89 mergers in the period 1981-98 and find a significant negative CAR of -1.01% for the period [-1, 1]. Yeh (2007) looks into 109 mergers and 36 tender offers during the years from 1981 to 1998. For the period [-1, 1] he finds a insignificant negative CAR of –0.34%, and for the period [-10, 1] a significant positive CAR of 1.44%. Kang et al. (2000) investigate bidder returns over the period 1977 to 1993. They confirm the positive cumulative abnormal return for the period [-1,0] found in previous research with a CAR at 1.17%

Figure 9-1 CAR results of previous event studies on Japanese mergers CAR -5-4-3-2-1AD12345 Bidder company Pettway & Yamada (1986) 1977-1984 0.70%

Ito (1989) 1971-1987 1.15%

Komoto (2002) 1980-1999 -2.10%

Yeh and Hoshino (2002) 1981-1998 -1.01%

Yeh (2007) * 1981-1998 -0.34%

Kang et al. (2000) 1977-1993 1.17%

Target company Pettway & Yamada (1986) 1977-1984 1.33%

-0.07%

-0.86%

Ito (1989) 1971-1987 1.26%

-2.85%

-1.75%

Komoto (2002) 1980-1999 -4.90% * Sample includes 109 mergers and 36 tender offers.

For target companies Pettway and Yamada (1986) find a significant positive abnormal return of 1.57% at [-1] and a significant negative abnormal return of -1.4% at [+1]. The resulting CARs for the periods [-1,0] and [-1,+1] are respectively positive at 1.33% and negative at -0.07%. The CAR for the longer interval [-5,+5] results in a negative CAR of - 0.86%. Ito (2000) finds similar results for the 31 target companies in his sample: the CAR for [-1, 0] is positive at 1.26%, but turns negative for the periods [-1,+1] and [-5,+5] at - 2.85% and -1.75%, respectively. Komoto (2002) finds a negative cumulative abnormal return of -4.9% for the period [-5,+5].

165

177 Chapter 9 Merger activity

U.S. mergers - Findings on abnormal returns for U.S. bidder companies show mixed results. While some research finds positive abnormal returns (Asquith 1983; Asquith, Bruner, and Mullins 1983; Loderer and Martin 1990), others find negative results (Langetieg 1978; Dodd 1980; Servaes 1991; Kaplan and Weisbach 1992). The general conclusion drawn by both Jensen and Ruback (1983) and Bruner (2002) is that the abnormal returns of U.S. bidder companies are on average zero and that bidders’ investors earn their required rate of return. The evidence on positive abnormal returns for U.S. target companies at the announcement date is considerable (e.g. Langetieg 1978; Servaes 1991; Kaplan and Weisbach 1992).

9-2-3 Profitability drivers Most empirical studies employ additional explanatory variables to examine the overall effect on stock prices, commonly addressed as profitability drivers. In this subsection we discuss the following drivers: (i) relative size of the target company, (ii) corporate diversification deals, (iii) unlisted versus listed target company, (iv) payment in cash versus stock, and (v) hostile versus friendly deal. Specifically for the Japanese market, earlier studies look into the importance of (vi) the main bank and (vii) membership of keiretsu groups.

Relative size - The relative size of the target company appears to have a positive effect on the abnormal return of the bidder company. Asquith et al. (1983) investigate the gains accruing to bidding companies’ shareholders in 214 merger bids during the period 1963-1979. They find that a merger bid for a target half the bidder's size produces an estimated cumulated abnormal return of 1.84% greater than a bid for a target that is one-tenth the bidder company's size. They do not find a relation between the abnormal returns of 54 listed target companies and the relative size of bidder and target. Servaes (1991) examines 704 completed takeovers in the period 1972-87. The research focuses on the abnormal returns from the announcement date of the takeover until the effective date or the delisting date. He finds that the size of the target relative to the bidder has a positive influence on the gains of the bidder company and the total gains, with the latter being significantly positive at 4.56%. The results also show that gains are higher when a good performing bidder company takes over a poorly performing target company. Pettway and Yamada (1986) look into the size effect for Japanese mergers, classifying a merger as “large” when the target company was more than 20% of the bidder company’s equity size. In contrast to the U.S., their tests show a significant and greater abnormal return for small mergers at the announcement date.

166

178 Section 9-2 Previous research

Corporate diversification - Corporate diversification is found to destroy value. Studies by Berger and Ofek (1995), Comment and Jarrel (1995), Lang and Stulz (1994) and Morck et al. (1990) support this finding. In contrast, Matsusaka (1993) and Hubbard and Palia (1999) find evidence that cross-industry M&A did create value in the U.S. in the 1960s. Morck et al. (1990) look into the announcement effect of unrelated diversification acquisitions for bidder companies’ shareholders in the period 1975-87. The sample consists of 91 unrelated acquisitions and 235 related acquisitions. They find that “[M]ean returns in related vs unrelated acquisitions are not statistically of substantively different in the 1970s, but are in the 1980s” (p.42). In the period 1980-87 the mean abnormal return for related acquisitions is 2.88%, and for unrelated acquisitions -4.09%. Matsusaka (1993) examines 199 acquisitions in the years 1968, 1971 and 1974. The sample consists of 67 diversification acquisitions and 132 related acquisitions. Shareholders of bidder companies receive a significantly positive abnormal return at the announcement of a diversification acquisition. Gains of bidder companies that made a related acquisition are significantly negative in the sample years. Hubbard and Palia (1999) investigate 229 diversification mergers and 163 related mergers between 1961 and 1970. They find positive abnormal returns in diversification mergers as well as related mergers for the shareholders of the bidder companies.

Cash vs stock - The abnormal returns for bidder companies and target companies in mergers paid by cash are higher than stock-for-stock mergers. Travlos (1987) examines the influence of the method of payment on the abnormal return of bidder companies. The sample consists of 167 successful takeovers (126 merger offers and 41 tender offers) in the period 1972 through 1981. The shareholders of the 60 bidder companies that financed the takeover through the exchange of common stock lost 1.47% in the two-day period [-1, 0]. The shareholders of the bidders that paid cash for the shares of the target company received normal rates of returns at the announcement. Heron and Lie (2002) investigate the influence of the method of payment on the abnormal return for the bidder company, the target company and the companies combined. Their sample consists of 859 acquisitions by 657 bidder companies, announced and completed between January 1985 and December 1997. For the 3-day interval [-1, +1] they find an average abnormal return for target companies at 25.4% for cash payments and at 17.1% for stock payments. For the bidder companies their results are similar to Travlos (1987). The shareholders get a normal return when the acquisition is paid with cash and negative at -1.9% when it is stock-financed. The combined result of acquisitions for cash payments is 5.3% and significantly higher than the 0.9% for the stock exchange.

Unlisted Target vs Listed Target - The announcement effect of bidder companies in mergers with unlisted target companies is stronger than with listed targets companies. Chang (1998) examines the shareholder wealth created by bidder companies in 281 merger offers of privately held targets. The sample covers the period 1981-92 and consists of 131 cash offers and 150 stock offers (one third of the offers are mixed offers of stock and cash). The abnormal return of the two-day announcement period [-1,0] is significantly positive at 2.64% when stock is offered, but there is no significant gain when cash is paid at 0.09%. In the matched sample of listed targets, stock payment results in a significant negative return at -2.64% and cash payment in an insignificant abnormal return of -0.02%.

167

179 Chapter 9 Merger activity

Faccio et al. (2004) investigate almost 4,500 acquisitions by European companies over the period 1996 to 2001. Bidder companies that acquired unlisted target companies realized a significant excess return of 1.48% and bidder companies of listed target companies an average negative abnormal return of –0.38%. Kang et al. (2000) do not find a significant difference between the abnormal returns of bidder companies in mergers with privately or publicly held target companies in Japan.

Hostile versus friendly - Announcement of tender offers and hostile takeovers results in higher abnormal returns than friendly mergers for both bidder companies and target companies. Rau and Vermaelen (1998) investigate bidder companies’ long-term performance in mergers and tender offers, announced and completed in the period 1980-91. Performance is computed over three years after completion of the bid and compared with an equally weighted control portfolio. Bidder companies underperform the control portfolio in mergers and realize a small but significant gain in tender offers. The research finds a large difference in performance between value bidders (high book-to-market ratio) and glamour bidders (low book-to-market ratio). Glamour bidders perform worse than value bidders in mergers and tender offers (performance of glamour bidders in mergers is negative and significantly different from the positive abnormal return of value bidders). Lang, Stultz and Walking (1989) examine bidder, target and combined gains of 87 successful tender offers in the period 1968-86. The influence of the bidder’s and target’s Tobin’s q on the abnormal returns for the period [-5,+5] is investigated. On the other hand, Healy, Palepu and Ruback (1992) found that hostile deals were associated with insignificant improvements in cash flow returns, owing possibly to the payment of higher acquisition premiums.

Main bank - In the previous section we discussed monitoring by the main banks. The presence of the main bank allows companies to focus on long-term goals like business growth and market share, rather than short-term stock price maximization in the interest of shareholders (Nakatani 1984; Odagiri 1992; Kester 1991; Porter 1992; Kang and Stulz 1996). According to Aoki (1984), protecting access to future loans is often more important than shareholder wealth maximization for Japanese companies. Kester (1986) argues that managers have an incentive to maximize the value of the nexus of trading and financial contracts between companies, not necessarily shareholder wealth. Morck and Nakamura (1999) indicate that banks use their influence on boards primarily to maximize the value of their loan portfolios, which can deviate substantially from shareholder value maximization and from efficiency. Kang et al. (2000) look into the influence of the main bank on abnormal return of bidder companies in mergers. They find for the companies affiliated with a main bank a significant positive mean abnormal return of 1.2% for the period [-1,0]. For companies that are not affiliated with a main bank the mean is insignificant at 0.5% and the two groups are not significantly distinguishable from each other. Based on these findings Kang et al. (2000) argue that “[A]nnouncement returns display a strong positive association with the strength of acquirer’s relationships with banks” (p.2197).

Keiretsu - Pettway and Yamada (1986) look into the influence of keiretsu groups in Japanese domestic mergers by testing the effect of shareholdings on the abnormal return of

168

180 Section 9-3 Shareholder wealth of bidder companies

bidder companies. They distinguish mergers between affiliated companies (shareholdings of more than 50% in the target company), related companies (shareholdings between 20%- 50% in target), and unrelated. A merger with an unrelated company has a significant positive effect on abnormal return at the announcement date.

9-3 Shareholder wealth of bidder companies R Bidder companies show a positive abnormal return around the announcement date of approximately 1.4%. R We do not find an association between announcement returns and the strength of a bidder company’s relationship with a main bank. R Announcement returns are related to the presence of a common shareholder holding shares in both the bidder and the target company and whether the deal took place after 1997.

In this section we examine the impact of a merger on shareholder wealth of bidder companies. We also investigate volume data in our analysis to determine whether shareholders consider the announcement as an information event. Although the bidder company’s share price may not move around the announcement date, increased trading volume could. In particular, we examine the following: Firstly, we investigate the importance of the bidder company’s relationship to its main bank. As described in the previous chapters, the importance of the main bank is argued to have diminished in recent years. Our dataset contains 136 merger cases that took place in the period 1993-2003 and allows us to test this hypothesis. Kang et al. (2000) find a significant positive association of the main bank with announcement returns for the period 1977-1993. Secondly, we investigate the influence of keiretsu relations. We look into the horizontal keiretsu groups and shareholdings. We examine the influence on the stock price of a bidder’s shareholding percentage in the target prior to the merger. In case a bidder already owns a controlling stake in a target company, the final acquisition is less likely to have a significant effect. We also look into the influence of a common shareholder that holds shares in both the bidder company and the target company. Thirdly, we examine how other profitability drivers affect bidder companies’ shareholder wealth. We study the influence of the relative size of the target company, the difference between listed and unlisted targets, industry-relatedness, a ‘rescue motive’ merger, and the financial structure and size of the bidder. We exclude mergers in the financial industry and regulated industries such as utilities and telecommunications. In general, we include very detailed information on listed as well as unlisted targets in our analysis, which to our knowledge has never been done before.

9-3-1 Data-set We look into domestic mergers between non-financial companies in the 11-year period from 1 January 1993 until 31 December 2003. Our sample consists of bidder companies listed on the First Section or the Second Section of the Tokyo Stock Exchange (TSE), acquiring either a listed or an unlisted Japanese target company. We investigate all press articles from 1993 to 2003 related to mergers in Japan from the Nihon Keizai Shimbun (Japan Economic Journal), Nikkei Sangyǀ Shimbun (Industrial Journal), Nikkei Ryǌtǌ

169

181 Chapter 9 Merger activity

Shimbun (Distribution Journal) and Nikkei Kinyǌ Shimbun (Finance Journal). From these press articles we collect the initial public Announcement Date (AD) and the Effective Date (ED) of the merger. The AD is defined as the day the merger announcement appears in the press for the first time. To eliminate cases in which the bidder company already has full ownership of the target before the announcement date, we examine various issues of Nikkei Kaisha Jǀhǀ (Nikkei Company Information) and Kigyǀ Keiretsu Soran to collect data on the biggest 10 shareholders for listed companies. For unlisted companies we refer to the publication Nikkei Kaisha Sokan (Nikkei; Annual Corporation Reports (Unlisted)) that shows data on the company’s largest shareholders and accounting information. We only include unlisted companies for which information on shareholders in the year of the merger or the year prior to the merger is available. If target companies are fully owned by the bidder, they are dropped from the sample. Table 9-1 indicates that our final sample consists of 136 merger cases, in which Japanese bidder companies merged with 69 listed targets and 67 unlisted target companies. The percentage of deals between a listed bidder and unlisted target company in our sample is lower than in other research. The sample of Kang et al. (2000), for example, consists for 70% of mergers with an unlisted target. One reason for this difference may be our requirement of publicly available information on shareholder data for unlisted companies.

Table 9-1 Descriptive statistics Total Company Shareholdings Common SHH Year Cases % total LB< LB&UT Total B in T T in B Cross All > 5% 1993118%65 7700 53 1994107%37 5500 31 199543%13 3300 11 1996118%47 3201 42 199797%63 5500 32 1998 18 13% 10 8 8 4 1 3 5 4 1999 20 15% 13 7 11 11 0 0 8 3 20001511%96 6600 97 200197%27 2101 43 20021410%95 4400 73 20031511%69 5500 55 Total 136 100% 69 67 59 53 1 5 54 34 % total 51% 49% 43% 39% 1% 4% 40% 25% Period 1993-1997 Cases 45 20 25 23 22 0 1 16 13 1998-2003 Cases 91 49 42 36 31 1 4 38 21 1993-1997 % total 33% 15% 18% 17% 16% 0% 1% 12% 10% 1998-2003 % total 67% 36% 31% 26% 23% 1% 3% 28% 15% Notes The sample consists of 136 Japanese bidders listed on the Tokyo Stock Exchange for which the announcement date of the merger is between 1 January 1993 and 31 December 2003. The AD is defined as the day that the merger announcement appears in the press for the first time. The merger is included in the sample if sufficient accounting information can be retrieved on unlisted target companies. For 'Company': LB = Listed Bidder, LT = Listed Target, UT = Unlisted Target. For 'Shareholdings': B = Bidder, T= Target, Cross = Cross Shareholdings. Sources Shareholder information: Nikkei Kaisha Jǀhǀ (Nikkei Company Information) and Kigyǀ Keiretsu Soran is used for listed companies. Nikkei Kaisha Sokan (Nikkei; Annual Corporation Reports (Unlisted)) is referred to for unlisted companies.

Table 9-1 shows the number of merger cases in each year of our sample period, and confirms the general trend in M&A activity in Japan with a considerable increase after

170

182 Section 9-3 Shareholder wealth of bidder companies

1997. We find the highest percentage, at 15% of all merger cases, in 1999. Between the years 1993 and 1997 our sample contains nine observations annually and this number increases to fifteen in the period 1998-2003. In the latter period the number of mergers with a listed target is slightly higher. The number of transactions in which there were shareholdings between the bidder company and the target company at AD is also given in table 9-1. We find shareholdings among the merging companies in 59 cases of our total sample (43%). Most important in these shareholdings is ownership of the target company by the bidder company. Including cross-shareholdings, the bidder company has shareholdings in 58 target companies, which is 42.6% of all mergers. Shareholdings by the target company in the bidder are small with only six merger cases, and the number of mergers with cross-shareholdings is only 4% of the total sample. We find that in 54 mergers there is a common shareholder with shareholdings in both merging companies. In 34 merger cases this common shareholder has ownership in both companies of more than over 5%, the percentage that needs to be reported according the Securities and Exchange Law. Panel A of table 9-2 summarizes transaction-specific details and shows that the bidder companies’ average shareholding in target companies was 16.1% for the total sample. Focusing only on bidder companies with shareholdings in the target company, this percentage increases to 37.8%, indicating a strong relationship with the target company for these bidders. The common shareholder has an average ownership in the bidder company and the target company of respectively 25.2% and 29.7%. 75 When we only look at shareholdings of more than 5% in the bidder and target, these shareholdings increase to 35.5% in the bidder and 42.4% in the target company. In our examination of the influence of the vertical groups, we focus on bidder company’s shareholdings in the target, and the shareholdings of the common shareholder with more than 5% ownership in the merging companies. In its role as parent company, the common shareholder may exert a strong influence over the merging decision of the bidder company and target company. We next examine the main bank system and define a main bank as the company’s most important lender that belongs to the company’s largest five shareholders. Following Aoki et al. (1994), who argues that “having the largest share of short-term loans, rather than the size of the overall loan share, is thought to be the hallmark of the main bank” (p.8), we look for the bank that is the largest lender of short-term loans. We take the largest loan provider as the most important lender (similar to Kang et al. 2000), if the company does not have short-term loans with a financial institution. We could identify a most important lender based on short-term loans for 106 bidder companies and for another 10 bidder companies based on total loans. This gives a total of most important lender relations for 116 bidder companies (85% of our total sample). To investigate the strength of the relationship, we look into the shareholding characteristic of the main bank system. We find that in 116 bidder companies with a most important lender, there is no shareholding relationship in 27 cases. The most important lender belongs to the largest six to ten shareholders in fourteen cases, with an average shareholding of 1.76%. Shareholdings of the main bank in the company can have an important influence on mergers of Japanese companies. We define a bank as a company’s main bank when it is the company’s most important lender and belongs to the company’s largest five shareholders. For 75 bidder

75 This includes the 20 merger cases in which the large common shareholder has shareholdings lower than 5% in both companies.

171

183 Chapter 9 Merger activity

companies the most important lender was one of the largest five shareholders (55.1% of the sample) with shareholdings at a mean of 4.18% and median of 4.5%. Following Nakatani (1984) and Kang et al. (2000), we use various issues of Keiretsu no Kenkyǌ to investigate whether a bidder company is part of one of the six horizontal keiretsu, Mitsubishi, Mitsui, Sumitomo, DKB, Fuji and Sanwa.76 We identify 68 bidder companies that are related to one of these groups (50% of our sample). Only 48 of the 75 bidder companies with a main bank (64%) are part of a horizontal keiretsu. Hence, the percentage of bidder companies with a main bank is for members of a horizontal keiretsu 71%, and for independent bidder companies 40%. This is an interesting finding because an often-used characteristic for Japanese corporate governance is that, although a Japanese company might be independent of a horizontal keiretsu, it is likely to have relations with a main bank. In our sample we find that the reverse is also true: members of a horizontal keiretsu do not always have a main bank. Based on the merger motives given during the press conference at AD, we investigate whether a merger was made by the bidder with a rescue-motive. This should, however, be interpreted with care because the reasons put forward during a press conference might not reflect the real motivation for a merger. The actual number of rescue-mergers is likely to be higher than the 11 merger cases identified (8.1% of total sample). In all rescue merger cases the bidder company has a main bank and in seven cases the bidder company belongs to a horizontal keiretsu. These findings suggest that being part of a horizontal keiretsu or having a main bank, has an influence on a bidder company for engaging in a rescue merger. We further find that most rescue mergers took place in the first part of our sample period, with ten in or before 1998 and one in 2000. Panel B of table 9-2 provides some financial data of the bidder and the target companies in our sample. The mean of total assets and total sales is for bidder companies 295 billion yen and 445 billion yen, respectively. This is considerably higher than target companies with mean total assets of 68 billion yen and mean total sales of 89 billion yen. The panel indicates there is a large difference in size between listed and unlisted target companies. Total assets and total turnover of listed targets and unlisted targets are respectively 32% and 10% of the bidders. When looking at the amount of total debt, these percentages are slightly higher but similar. The leverage ratio, measured as total debt to total assets, is approximately the same for the bidder company and the listed target company at 46.8% and 52.8% respectively. The unlisted target, on the other hand, shows an extremely high leverage ratio of 75.2%, which raises the average leverage percentage for target companies to 62.4%. Leverage, measured as total debt divided by the sum of the book value of debt and market value of equity, results in a mean and median for bidder companies at respectively 41.2% and 42.4%. The leverage-ratio for total loans is 22.2% and shows that bidder companies use loans for about half of their total debt. For bidder companies with a main bank we also measure leverage as main bank loans divided by the sum of the book value debt and market value equity, and find a ratio of 6.8%. So, a relative large percentage of outstanding loans is from the main bank.

76 Since Keiretsu no Kenkyuu was published up to 2000, we used this latest issue to classify the horizontal keiretsu in the subsequent three years.

172

184 Section 9-3 Shareholder wealth of bidder companies

Table 9-2 Summary Merger Specific and Company Specific Statistics Panel A : Merger Specific Variable Mean Median St.Dev. Fraction bidders with main bank 0.551 Fraction bidders part of keiretsu 0.500 Fraction of privately held targets 0.493 Fraction of intra-industry mergers 0.713 Fraction of rescue mergers 0.081 Target total assets / bidder total assets 0.521 0.314 0.654 Bidder shareholdings in target (%): All bidders (n=136) 16.1 0.0 23.9 Bidders with shareholdings (n=58) 37.8 37.1 22.7 Common shareholder shareholdings (%): Shareholdings in bidder (n=54) 25.2 22.3 19.9 Shareholdings in target (n=54) 29.7 20.1 28.7 Common shareholder shareholdings of 5% (*) (%): Shareholdings in bidder (n=34) 35.5 33.7 17.5 Shareholdings in target (n=34) 42.4 37.6 28.2 PanelB:CompanySpecific Bidder Target Variable Mean Median St.Dev. Mean Median St.Dev. Total assets (billions of yen) 295 51 620 All 68 19 130 Listed 98 36 153 Unlisted 35 10 86 Total sales (billions of yen) 445 103 1,016 All 89 26 155 Listed 139 77 196 Unlisted 37 15 65 Total debt (billions of yen) 171 17 334 All 48 9 102 Listed 62 14 110 Unlisted 30 7 88 Total loans (billions of yen) 56 9 106 Market value listed companies (bln yen) 216 38 612 60 17 145 Relative market cap 0.484 0.333 0.670 Leverage: Total debt/Total Assets 0.468 0.521 0.304 All 0.624 0.680 0.320 Listed 0.528 0.509 0.350 Unlisted 0.752 0.757 0.224 Total debt/(bv debt + mv equity) 0.412 0.424 0.275 Listed 0.496 0.476 0.275 Total loans/(bv debt + mv equity) 0.222 0.165 0.220 Main bank loan/(bv debt + mv equity) 0.068 0.042 0.079 Notes Common shareholder has shareholdings of 5% or higher in bidder and target company. The sample consists of 136 Japanese bidders listed on the Tokyo Stock Exchange for which the announcement date of the merger is between 1 January 1993 and 31 December 2003. The AD is defined as the day that the merger announcement appears in the press for the first time. The merger is included in the sample if sufficient accounting information can be retrieved on unlisted target companies. Shareholder information: Nikkei Kaisha Jouhǀ (Nikkei Company Information) and Kigyǀ Keiretsu Soran is used for listed companies. Nikkei Kaisha Sokan (Nikkei; Annual Corporation Reports (Unlisted)) is referred to for unlisted companies. The main bank is defined as a company’s most important lender which belongs to the company’s largest 5 shareholders. Keiretsu no Kenkyǌ is used to investigate whether a bidder company is part of one of the six horizontal keiretsu. Accounting data is retrieved from the Nikkei Needs Tapes.

Following Odagiri and Hase (1989), we make an inter-industry matrix based on the Nikkei Needs industry classification. Figure 9-2 shows the concentration of mergers on the diagonal line, which indicates that most mergers in our sample are between companies in

173

185 Chapter 9 Merger activity

the same industry. Intra-industry merger cases total 97 cases (71.3%) and inter-industry total 39 cases (28.7%). Bidder companies are most active in the trading sector (20 mergers), construction industry (fourteen mergers) and miscellaneous services industry (thirteen mergers). Resulting from the primacy of intra-industry mergers, we see that most companies were also taken over in the trading sector, construction industry, and miscellaneous services industry with respectively 26, twelve and fifteen mergers. Bidder companies that merged with target companies from other industries are primarily from the machinery industry (five mergers) and the trading sector (four mergers). Target companies taken over by bidder companies from other industries are concentrated in the trading sector (ten mergers) and the transportation equipments and real estate industry (five mergers).

Bidder Target 1 3 5 7 9 11131517192123252729313335374143455253555759616365676971Total

15 5 31 1 53 14 7161 1 9 911 2 11 2 2 13 1 1 2 15 9 9 17 1 1 19 1 1 4 1 7 21 117 9 23 3 3 25 0 27 2 2 29 1 1 1 1 1 5 31 1 1 33 1 2 1 4 35 0 37 0 41 12 12 43 1 4 2 1 16 1 1 26 45 77 52 0 53 1 1 3 5 55 0 57 11 59 33 61 0 63 0 65 11 67 0 69 0 71 1 1 2 11 15 Total53310222946126040230014208003030110013136 Figure 9-2 Inter-industry vs. intra-industry deals Notes The sample consists of 136 Japanese bidders listed on the Tokyo Stock Exchange for which the announcement date of the merger is between 1 January 1993 and 31 December 2003. The AD is defined as the day that the merger announcement appears in the press for the first time. The merger is included in the sample if sufficient accounting information can be retrieved on unlisted target companies. The table presents an inter-industry matrix based on the Nikkei Needs industry classification. For the unlisted companies we also used the classification given by Nikkei Needs in the respective publication.

9-3-2 Results Univariate Analysis - Abnormal returns are computed using the event-study methodology, following Dodd and Warner (1983) 77 . Our test period starts 20 days prior to the announcement date and stops 90 days after. We determine normal returns by calculating betas using daily return data over the 200 trading days preceding the test period, using the TOPIX index as our benchmark. The daily abnormal return is compounded over various time intervals to get the cumulative average abnormal return (CAR). We use standard t-

77 Please refer to Appendix I of this chapter.

174

186 Section 9-3 Shareholder wealth of bidder companies

statistics to test the hypothesis that the average CARs are equal to zero. To determine 'normal trading volume', we take the average of the ratio between daily trading volume in the bidder's stock and the total daily trading volume in all TOPIX stocks over the 200 days until 20 days before the announcement. We define 'abnormal volume' during the test period as the difference between the ratio of actual daily trading volume in the stock and all TOPIX stocks on the one hand, and the normal volume on the other. Figure 9-3a shows the effect of the announcement on bidder's stock prices. The stock price starts rising 2 days before the deal is announced and peaks on the announcement day. Thereafter, the price quickly falls and the overall CAR becomes statistically insignificant. This finding indicates significant slippage before the announcement is made. Trading volume does not rise until the actual announcement day and stays high until three days after the announcement. It returns to pre-announcement levels after five to seven days.

2.5% 1.00 3.0 1.00 Excess trading volume median 2.5 2.0% p-value 0.80 0.80 2.0 1.5%

1.5 0.60 0.60 1.0%

1.0 0.5% 0.40 0.40 0.5

0.0% 0.0 0.20 0.20 CAR (-10 to +10) -0.5% median -0.5 p-value (rhs)

-1.0% 0.00 -1.0 0.00 -10-9-8-7-6-5-4-3-2-10 1 2 3 4 5 6 7 8 910 -20-18-16-14-12-10-8-6-4-202468101214161820 Figure 9-3a CARs Japanese bidders Figure 9-3b Excess trading volume Japanese bidders Notes The sample consists of 136 Japanese bidders listed on the Tokyo Stock Exchange for which the announcement date of the merger is between 1 January 1993 and 31 December 2003. The AD is defined as the day that the merger announcement appears in the press for the first time. The merger is included in the sample if sufficient accounting information can be retrieved on unlisted target companies. Cumulative Abnormal Returns (CARs) are computed from a market model estimated from days -220 to -20 relative to the announcement. Excess trading volume is determined by taking the average of the ratio of daily trading volume in each stock and daily trading volume in the TOPIX index over the period from -220 to -20 days before the announcement. The actual trading volume over the test period is divided by this 'normal' ratio. P-values indicate whether average excess trading volume ratio differs significantly from 1.

175

187 Chapter 9 Merger activity

Table 9-3 CARs and Excess Volume for Japanese bidders Panel A: CARs around the announcement date Interval # obs. Mean CAR Median CAR 1.37% 0.46% AD-1 to AD 136 (0.01) (0.08) 0.57% -0.01% AD-1 to AD+1 136 (0.45) (1.00) 0.87% 0.00% AD-5 to AD+5 136 (0.33) (1.00) 4.61% 3.82% AD-5 to AD+90 106 (0.03) (0.03) -0.36% 0.00% ED-1 to ED 136 (0.43) (0.64) Panel B: Excess Volume around the announcement date Interval # obs. Mean EV Median EV 2.51 1.45 AD to AD+1 127 (0.00) (0.00) Notes The sample consists of 136 Japanese bidders listed on the Tokyo Stock Exchange for which the announcement date of the merger is between 1 January 1993 and 31 December 2003. The AD is defined as the day that the merger announcement appears in the press for the first time. The merger is included in the sample if sufficient accounting information can be retrieved on unlisted target companies. Cumulative abnormal returns (CARs) are computed from a market model estimated from days -220 to -20 relative to the announcement. ED denotes the effective date of the merger. P-values for t-tests are reported in parentheses below the mean, and p-values from sign-rank tests are reported in parentheses below the median. The sample in Panel B consists of 127 Japanese bidders. Excess Volume (EV) is calculated by taking the ratio between actual trading volume and ‘normal trading volume’. To determine 'normal trading volume', we take the average of the ratio between daily trading volume in the bidder's stock and the total daily trading volume in all TOPIX stocks over the 200 days until 20 days before the announcement. P-values for t-tests, whether values differ significantly from 1, are reported in parentheses below the mean, and p-values from sign-rank tests are reported in parentheses below the median.

Panel A of table 9-3 shows the mean and median CARs around the announcement date (AD) and the effective date (ED) for the 136 bidder companies in our sample. The mean CAR [-1,0] is 1.37 % and significant at the 0.01 level. The sign-rank test indicates that the median CAR [-1,0] is significantly different from zero at the 0.10 level. When we expand the test-window to [-1,+1] and [-5,+5], the CAR remains positive but is not significantly different from zero. For the longer time-window of CAR [-5,+90] the mean is 4.61% and statistically significant. This indicates a positive development in abnormal returns during the period after the merger announcement. At the effective date, on the other hand, the CAR [ED-1,ED] is negative. Our overall results indicate a positive market valuation on the merger announcement of bidder companies around AD and are in line with previous research on Japan. Our findings of the expanded windows support the earlier findings of a positive development by Kang et al. (2000) and Pettway and Yamada (1986). Panel B in table 9-3 shows that Excess Volume (EV) increases sharply on the announcement date and remains high on the day after. Overall, trading volume is on average 2.51 times (median: 1.45 times) the ‘normal’ level on the announcement and following day. The number of companies in the sample falls to 127, because no trading volume was registered for nine companies around the announcement date. In the following tests, we take the average EV of the announcement day and the day after that [0,+1].

Main bank and keiretsu linkages Table 9-4 categorizes the sample according to some key characteristics of the bidder company and the transaction, and compares CAR [-1,0] across these characteristics. We first look into the importance of the main bank in explaining abnormal returns. We

176

188 Section 9-3 Shareholder wealth of bidder companies

consider the main bank’s shareholding of a company and its provision of loans to the company. In short, we find no evidence that having relations with a main bank has a significantly more positive influence on shareholder value. We first separate bidder companies according to whether or not having a main bank defined as the largest lender that belongs to the largest 5 shareholders. The mean CAR for bidders unrelated to a main bank is 1.95% and statistically significant at the 0.01 level. The 75 bidders that are related to a main bank have a CAR of 0.90%, insignificantly different from zero. We do not find noticeable differences between the two groups. Bidder companies with below median main bank leverage, measured as main bank loans divided by the sum of book value debt and market value equity, have statistically significant mean and median CARs at respectively 2.08% and 0.83%. Bidder companies with main bank leverage above the median have a mean CAR at 0.66% and median CAR at 0.12%, both being insignificant. The medians of the two groups are distinguishable at a 0.10 level. Panel A of table 9-5 shows that the presence of a main bank does not have any effect on excess trading volume either. In other words, the stock market clearly views the announcement as information, but whether a bidder has an affiliation with a main bank or not, does not have any influence.

177

189 Chapter 9 Merger activity

Table 9-4 CAR [-1, 0] for Japanese bidders categorized by characteristics of bidder and transaction Mann- Number Mean CAR Median CAR T-Test Whitney of obs (p-value) (p-value) (p-value) (p-value) All Cases 136 1.37% 0.46% (0.011) (0.085) Panel A : Main Bank and keiretsu linkages Bidderisaffiliatedwithmainbank 75 0.90% 0.15% -0.982 (0.236) (0.489) (0.328) Bidder is unaffiliated with main bank 61 1.95% 0.82% -1.144 (0.010) (0.093) (0.253) Bidder’s main bank loan ratio is above 68 0.66% 0.12% -1.335 sample median (0.412) (0.545) (0.184) Bidder’s main bank loan ratio is below 68 2.08% 0.83% -1.519 sample median (0.004) (0.086) (0.129) Bidder belongs to horizontal keiretsu 68 1.00% 0.42% -0.693 (0.197) (0.114) (0.490) Bidder independent of horizontal keiretsu 68 1.74% 0.46% -0.453 (0.021) (0.464) (0.651) Panel B : Other factors Bidder’s bank loan ratio is above sample 68 0.71% -0.05% -1.244 median (0.412) (0.904) (0.216) Bidder’s bank loan ratio is below sample 68 2.03% 1.26% -2.194 median (0.002) (0.007) (0.028) Bidder’s leverage ratio is above sample 68 0.83% 0.10% -1.027 median (0.232) (0.904) (0.307) Bidder’s leverage ratio is below sample 68 1.92% 0.55% -1.375 median (0.021) (0.027) (0.169) Bidder has shareholding in target 58 0.52% 0.09% -1.382 (0.499) (0.896) (0.169) Bidder does not have shareholding in 78 2.00% 0.92% -1.575 target (0.007) (0.040) (0.115) Bidder's shareholding in target is above 29 0.24% 0.20% -1.114 median (*) (0.728) (0.711) (0.267) Bidder's shareholding in target is below 107 1.68% 0.48% -0.417 median (*) (0.011) (0.098) (0.677) Bidder and target have common 54 2.21% 0.13% 1.282 shareholder (0.030) (0.684) (0.202) Bidder and target do not have common 82 0.82% 0.47% 0.062 shareholder (0.167) (0.075) (0.950) Common shh with 5% or more in bidder 34 3.32% 1.70% 2.146 and target (0.024) (0.392) (0.034) Common shh without 5% or more in 102 0.72% 0.44% 1.216 bidder and target (0.170) (0.163) (0.224)

178

190 Section 9-3 Shareholder wealth of bidder companies

Table 9-4 continued Mann- Number Mean CAR Median CAR T-Test Whitney of obs (p-value) (p-value) (p-value) (p-value) Relative size above sample median 65 2.63% 0.60% 2.330 (0.007) (0.103) (0.021) Relative size below sample median 65 0.17% 0.20% 1.572 (0.722) (0.457) (0.116) Acquisition motivated by rescue purposes 11 -2.45% -0.63% -2.164 (0.032) (0.549) (0.032) Acquisition not motivated by rescue 125 1.71% 0.49% -1.999 purposes (0.002) (0.038) (0.046) Merger in period 1993-1997 45 -0.03% 0.04% -1.874 (0.964) (1.000) (0.063) Merger in period 1998-2003 91 2.07% 0.54% -1.482 (0.005) (0.045) (0.138) Large bidders (market cap is above sample 68 1.05% 0.54% -0.610 median) (0.118) (0.114) (0.543) Small bidders (market cap is below sample 68 1.70% 0.44% 0.113 median) (0.046) (0.464) (0.910) Target is listed 70 1.17% 0.19% -0.395 (0.129) (0.550) (0.694) Target is unlisted 66 1.59% 0.52% -0.980 (0.037) (0.082) (0.327) Bidder and target are in same industry 97 1.44% 0.49% 0.211 (0.014) (0.125) (0.833) Bidder and target in different industries 39 1.19% 0.20% 0.431 (0.320) (0.522) (0.667) Notes The sample consists of 136 Japanese bidders listed on the Tokyo Stock Exchange for which the announcement date of the merger is between 1 January 1993 and 31 December 2003. AD denotes the initial announcement date, i.e. the first day on which the information related to the announcement was public before the end of the trading day. The AD is defined as the day that the merger announcement appears in the press for the first time. The merger is included in the sample if sufficient information can be retrieved on unlisted target companies. Cumulative abnormal returns (CARs) are computed from a market model estimated from days -220 to -20 relative to the announcement. Leverage ratio is computed as the ratio of the book value of total debt to the sum of the book value of debt and the market value of equity. Bank loan ratio is computed as the ratio of the book value of debt from financial institutions to the sum of the book value of debt and the market value of equity. Main bank loan ratio is computed as the ratio of the book value of debt from the main bank to the sum of the book value of debt and the market value of equity. P-values for t-tests are reported in parentheses below the mean, and p-values from sign-rank tests are reported in parentheses below the median.

179

191 Chapter 9 Merger activity

Table 9-5 EV [0, +1] for Japanese bidders categorized by characteristics of bidder and transaction Mann- Number Mean CAR Median CAR T-Test Whitney of obs (p-value) (p-value) (p-value) (p-value) All Cases 127 2.514 1.450 (0.000) (0.001) Panel A : Main Bank and keiretsu linkages Bidder is affiliated with main bank 73 2.554 1.420 0.182 (0.000) (0.101) (0.856) Bidder is unaffiliated with main bank 54 2.460 1.585 -1.095 (0.000) (0.002) (0.274) Bidder’s main bank loan ratio is above 66 2.605 1.445 0.376 sample median (0.000) (0.175) (0.707) Bidder’s main bank loan ratio is below 61 2.415 1.450 -1.071 sample median (0.000) (0.001) (0.284) Bidder belongs to horizontal keiretsu 67 1.800 1.290 -3.088 (0.000) (0.268) (0.003) Bidder independent of horizontal keiretsu 60 3.311 1.820 -2.465 (0.000) (0.000) (0.014) Panel B : Other factors Bidder’s bank loan ratio is above sample 61 2.074 1.540 -1.689 median (0.000) (0.004) (0.094) Bidder’s bank loan ratio is below sample 66 2.921 1.335 0.376 median (0.000) (0.082) (0.707) Bidder’s leverage ratio is above sample 64 2.143 1.495 -1.489 median (0.000) (0.023) (0.139) Bidder’s leverage ratio is below sample 63 2.891 1.360 -0.366 median (0.000) (0.023) (0.714) Bidder has shareholding in target 55 2.178 1.440 -1.164 (0.000) (0.177) (0.247) Bidder does not have shareholding in 72 2.770 1.495 -1.253 Target (0.000) (0.002) (0.210) Bidder's shareholding in target is above 28 1.744 1.335 -1.634 median (*) (0.017) (0.345) (0.105) Bidder's shareholding in target is below 99 2.732 1.510 -1.361 median (*) (0.000) (0.002) (0.174) Bidder and target have common 52 2.634 1.345 0.395 shareholder (0.000) (0.262) (0.693) Bidder and target do not have common 75 2.431 1.540 -0.439 shareholder (0.000) (0.001) (0.661) Common shareholder has shareholdings of 37 3.066 1.440 1.408 5% or more in bidder and target (0.000) (0.405) (0.162) Common shareholder does not have 90 2.287 1.460 0.470 shareholdings of 5% or more in bidder and (0.000) (0.001) (0.639) Relative size above sample median 58 3.359 1.740 3.276 (0.000) (0.001) (0.001) Relative size below sample median 63 1.705 1.320 2.719 (0.001) (0.450) (0.007) Acquisition motivated by rescue purposes 11 3.128 2.160 0.749 (0.060) (0.227) (0.456) Acquisition not motivated by rescue 116 2.456 1.440 1.140 purposes (0.000) (0.003) (0.254)

180

192 Section 9-3 Shareholder wealth of bidder companies

Table 9-5 continued Mann- Number Mean CAR Median CAR T-Test Whitney of obs (p-value) (p-value) (p-value) (p-value) Merger in period 1993-1997 65 2.619 1.440 0.426 (0.000) (0.033) (0.671) Merger in period 1998-2003 62 2.404 1.460 0.135 (0.000) (0.015) (0.893) Large bidders (market cap is above sample 67 2.322 1.440 -0.803 median) (0.000) (0.086) (0.423) Small bidders (market cap is below sample 60 2.728 1.510 -0.734 median) (0.000) (0.004) (0.463) Target is listed 65 2.394 1.330 -0.483 (0.000) (0.169) (0.630) Target is unlisted 62 2.639 1.545 -0.736 (0.000) (0.001) (0.462) Bidder and target are in same industry 89 2.533 1.440 0.116 (0.000) (0.014) (0.908) Bidder and target in different industries 38 2.469 1.600 -0.461 (0.000) (0.034) (0.645) Notes The sample consists of 136 Japanese bidders listed on the Tokyo Stock Exchange for which the announcement date of the merger is between 1 January 1993 and 31 December 2003. Because no trading volume was registered for 9 companies around the announcement date, the sample for Excess Volume falls to 127. AD denotes the initial announcement date, i.e. the first day on which the information related to the announcement was public before the end of the trading day. The AD is defined as the day that the merger announcement appears in the press for the first time. Excess trading volume is determined by taking the average of the ratio of daily trading volume in each stock and daily trading volume in the TOPIX index over the period from -220 to -20 days before the announcement. The actual trading volume over the test period is divided by this 'normal' ratio. P-values indicate whether average excess trading volume ratio differs significantly from 1. Leverage ratio is computed as the ratio of the book value of total debt to the sum of the book value of debt and the market value of equity. Bank loan ratio is computed as the ratio of the book value of debt from financial institutions to the sum of the book value of debt and the market value of equity. Main bank loan ratio is computed as the ratio of the book value of debt from the main bank to the sum of the book value of debt and the market value of equity. P-values for t-tests, whether values differ significantly from 1, are reported in parentheses below the mean, and p- values from sign-rank tests are reported in parentheses below the median.

Other factors Looking further into the influence of leverage on CARs, we find significant mean CARs for bidders with below median total leverage at 1.92% and below median bank loan leverage at 2.03%. Bidders with above median leverage do not have CARs significantly different from zero, implying that bank loans do not increase value at the time of a transaction announcement. Again, this is in contrast to the findings by Kang et al. (2000). The influence of leverage on trading volume is limited. We only find a significantly lower average trading volume for bidder companies with a bank loan ratio above the sample median. Pettway and Yamada (1986) find that not having shareholdings in the target by the bidder company has a positive effect on abnormal returns. We examine the shareholdings of the bidder in the target company and also the shareholdings of a common shareholder. In our tests on the influence of the bidder company’s shareholdings in the target company, the results confirm the findings by Pettway and Yamada (1986). The mean CAR for bidder companies without shareholdings are 2.00% and significant at the 0.01 level. This is, however, statistically indistinguishable from the 0.52% mean CAR of bidder companies that do own shares of the target. In an additional test we compare two groups based on the median of the bidder companies with shareholdings and find similar results. The common shareholder has a stronger influence on CAR. Bidder companies with a

181

193 Chapter 9 Merger activity

common shareholder have a mean CAR of 2.21%. It remains positive for the merger cases without a common shareholder but the CAR is lower and insignificant at 0.82%. If we take out the influence of smaller shareholdings, we find that in mergers with a common shareholder owning more than 5% of both companies, the CAR increases strongly to 3.32% at a p-level of 0.05. The CAR for bidder companies without a common shareholder decreases slightly to 0.72%, and the two groups are statistically different from each other. Hence, the influence of a parent company with shareholdings in the bidder and the target appears to be more important than the shareholdings of the bidder in the target company. While in U.S. research evidence is found that a larger target has a more positive effect on abnormal returns than smaller targets, Pettway and Yamada (1986) find the opposite for Japanese takeovers. The influence of the size of the target is measured by dividing the target companies’ total assets by the bidder companies’ total assets and separating the sample into groups above and below the median. We find a statistically significant CAR of 2.63% for large target transactions, which is significantly different at the 0.05 level from the CAR of small target transactions with an insignificant CAR of 0.17%. So, in contrast to Pettway and Yamada and in line with U.S. research, we find a positive influence of the target’s size on the bidder’s CAR. The effect on trading volume appears to be consistent with this finding; when the deal is large relative to the size of the bidder, excess trading volume is higher. We continue by investigating the influence of the ‘rescue’-motive given by the bidder company at the announcement date. This announcement results in a large negative CAR of -2.45%, compared to a significant CAR of 1.71% when it is not termed as a rescue-merger by the bidder. This difference is statistically significant. The latter half of the 1990s, when the legislative and accounting changes were announced and implemented, appears to have a strong positive influence over CAR. We compare various periods in our sample and find significant differences in CARs between the periods 1993-97 and 1998-2003, 1993-98 and 1999-2003, and 1993-99 and 2000-2003. The bidder companies’ mean CAR for the period 1998-2003 is 2.07% and significant at the 0.01 level. It is significantly different from the insignificant mean CAR of -0.03% in the period 1993-1997. The most important difference between the mergers in these periods is that the percentage of bidder companies affiliated with a main bank, which is considerably lower in the latter half of the 1990s (45% compared to 76% in 1993-97). We further find that in the latter period the average shareholdings of the common shareholder are higher and the number of mergers with a rescue motive is lower (3% versus 18% in period 1993- 97). We examine the influence of the bidder company’s size on CARs by dividing the sample into groups with the bidder’s market capitalization being above and below the median. We find a significant 1.70% mean CAR for smaller bidders, which is indistinguishable from the above median group of 1.05%. We also investigate whether, as Chang (1998) and Faccio et al. (2004) find, there is a positive influence on CAR when unlisted target companies are acquired. We find a significant CAR of 1.59% for unlisted targets, compared to an insignificant CAR of 1.17% for listed targets. The two groups are indistinguishable from each other, confirming the findings by Kang et al. (2000). Lastly, we look into the influence of the intra-industry and inter-industry mergers on CAR. In line with Kang et al. (2000), we find a significant and positive CAR for intra- industry mergers, but there is no difference between CARs of intra-industry and inter-

182

194 Section 9-3 Shareholder wealth of bidder companies

industry mergers.

Company and deal characteristics: CARs for mergers in progress We investigate the development of the abnormal returns in the 4.5 months after the announcement date as well. We examine the CAR [-5,+90], which incorporates the announcement effect as well as the stock market response to a merger bid in progress. Changes in the abnormal return after the announcement date indicate that new information becomes available and reflects the uncertainty at the announcement date and whether the merger was evaluated correctly. In order to avoid the influence of confounding events, we excluded cases where the effective date of the merger took place within 95 business days after the announcement date. This reduced our sample to 106 merger cases. The mean CAR [-5,+90] is 4.61% and significant at the 0.05 level. The sign-rank test indicates that the median of 3.82% is also significantly different from zero at the 0.05 level. However, because of the length of this period, returns may be affected by events unrelated to the merger announcement. We decided to exclude the 20 mergers with the largest fluctuation in abnormal returns. Both the mean and median of CAR [-5,+90] become insignificant in this smaller sample, at respectively 2.23% and 2.14%. Appendix II of this chapter shows the categorization of the smaller sample according the key characteristics of the bidder and the transaction conform table 9-4 for CAR [-1,0]. Although we only look at the 86 bidder companies, for comparison with the CAR [-1,0] results, the characteristics are based on the total sample (if for example the median is examined, we make a categorization based on the median of the total sample). Of all characteristics, only one significant difference between groups is found. The bidder’s average CAR for mergers in the period 1998-2003 is 5.03% and significant at the 0.1 level. It is statistically distinguishable from the CAR of -2.48% for mergers in the period 1993- 1997. This low number of significant differences can be attributed to the length of the period in which other information on the company becomes available, unrelated to the merger. Or, additional information related to the merger becomes available but is not in line with earlier market expectations. This paper does not allow us to examine this subject in detail, but we can conclude that for our sample no significant consistent additional information becomes available to the market in the 90 days after the announcement date.

Regression analysis CAR - Table 9-6 presents regressions with the cumulative abnormal returns CAR [-1,0] and the volume effect EV [-1,0] as dependent variables. Although we look into the total sample, for 6 target companies the total assets is not available, and the number of observations for CAR and volume effect is respectively 130 and 121 merger cases. The regressions control for the bidder company's size, the relative size of the transaction, bidder company’s ownership of the target company and the number of days between the announcement date and the effective date. We include a dummy variable that equals one if (i) the target company is listed, (ii) the bidder company belongs to a horizontal keiretsu, (iii) the merger has a rescue motive, (iv) it is an intra-industry merger, and (v) it takes place in or after 1998.

183

195 Chapter 9 Merger activity

Table 9-6 Regression analysis CARs and Excess Volume CAR EV Explanatory variables (1) (2) (3) (4) (5) (6) (7) (8) Bidder size (log of bidder total -0.002 0.006 -0.002 -0.002 -0.002 -0.001 -0.013 0.038 assets) (0.595) (0.384) (0.636) (0.596) (0.689) (0.748) (0.951) (0.903) Relative size of the transaction 0.008 0.033 0.008 0.008 0.008 0.008 1.002 2.291 (0.372) (0.055) (0.373) (0.376) (0.389) (0.398) (0.047) (0.008) Target is listed (dummy) 0.372 -0.037 0.002 0.002 -0.001 0.003 -0.296 -1.622 (0.877) (0.046) (0.889) (0.879) (0.966) (0.785) (0.615) (0.071) Rescue merger (dummy) -0.033 -0.060 -0.034 -0.034 -0.034 -0.033 1.178 0.563 (0.109) (0.031) (0.106) (0.110) (0.100) (0.104) (0.218) (0.673) Intra-industry transaction -0.003 -0.013 -0.004 -0.004 -0.003 -0.001 -0.019 0.261 (dummy) (0.782) (0.415) (0.769) (0.782) (0.805) (0.914) (0.975) (0.743) Bidder's ownership in target -0.006 0.001 -0.005 -0.006 -0.007 0.003 -0.510 -0.410 (0.827) (0.985) (0.842) (0.829) (0.792) (0.901) (0.671) (0.809) Bidder and target have common 0.019 0.049 0.019 0.019 0.186 0.927 shh ( > 5%) (0.158) (0.020) (0.155) (0.163) (0.776) (0.353) Common shh ownership in 0.065 bidder (0.058) Common shh ownership in target 0.066 (0.011) Bidder is member of horizontal -0.005 -0.011 -0.005 -0.005 -0.006 -0.009 -1.625 -2.967 keiretsu (dummy) (0.711) (0.554) (0.705) (0.716) (0.660) (0.511) (0.008) (0.002) Bidder is affiliated with main -0.003 -0.003 -0.003 -0.001 0.002 0.829 bank (dummy) (0.827) (0.792) (0.827) (0.907) (0.849) (0.163) Bidder'smainbankloanratio 0.109 -9.919 (0.337) (0.075) Bidder's bank loan ratio -0.010 (0.432) Leverage ratio (D/(D+Market cap 0.001 0.005 0.004 equity)) (0.981) (0.808) (0.857) Period (before 1998 and after) -0.010 -0.003 -0.010 -0.010 -0.010 -0.009 -0.687 -1.087 (0.402) (0.834) (0.432) (0.411) (0.421) (0.481) (0.236) (0.166) Number of days between AD and 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.013 ED (0.531) (0.193) (0.522) (0.533) (0.461) (0.485) (0.972) (0.072) Adjusted R2 0.024 0.142 0.017 0.016 0.030 0.054 0.082 0.197 Number of observations 130 73 130 130 130 130 121 71 Notes The sample used for this table consists of 136 companies (127 in the case of EV) listed on the Tokyo Stock Exchange for which the announcement date of the merger is between 1 January 1993 and 31 December 2003. AD denotes the initial announcement date, i.e. the first day on which the information related to the announcement was public before the end of the trading day. The AD is defined as the day that the merger announcement appears in the press for the first time. For 6 target companies the total assets is not available and the number of observations for CAR and EV is respectively 130 and 121 merger cases. A straightforward linear regression analysis was performed with a number of explanatory variables.

As a further investigation of the influence of the main bank system, we use a dummy variable in the regression that equals one if the bidder company is affiliated with a main bank. We also examine the importance of the percentage of main bank loans in the bidder company’s total assets. Leverage is investigated as the percentage of total loans and total debt in the bidder company’s total assets. Because the univariate analysis showed a strong influence on CAR for shareholdings of a common shareholder, we look into this characteristic in detail in the regression analysis as well. This is done with a dummy variable that equals one if there is a common shareholder with shareholdings of over 5% in the bidder and/or target. The ownership

184

196 Section 9-3 Shareholder wealth of bidder companies

percentages of the common shareholder in the bidder and the target are also examined separately.

In models (1) and (2) of table 9-6, we first examine the influence of the main bank. In model (1) we categorize the affiliation with a main bank by using the dummy variable. In model (2) we use the outstanding loan percentage with the main bank, qualifying companies without main bank loans as having no observation. So, in model (2) we only look at the CAR of bidder companies with a main bank and focus on the influence of the main bank loan percentage. In model (1) we do not find any significant relations between the explanatory variables and the dependent variable. For the main bank companies we see that a merger with a larger target is positively associated with CAR at a significance level of 0.1, confirming our univariate findings. In line with Chang (1998) and Faccio et al. (2004), we also find that a merger with a listed target company results in a bidder’s CAR to be lower than an unlisted target by 3.7%. Our previous findings for the negative influence of a rescue merger are also confirmed in the regression of the main bank companies. The presence of a common shareholder has a positive influence on the CAR at a significance level of 0.5. We execute additional tests with model (1) by increasing the limit of the shareholding percentage of the common shareholder. We find a significant relation of 2.3% when the common shareholder has shareholdings of 10% or more in the bidder and/or target. Thus, common shareholder’s shareholding above a certain threshold increases shareholder wealth. In models (3) and (4) we turn to the results on leverage by looking into the bidder company’s bank loan ratio and leverage ratio. The ratios of these two measures of leverage do not appear to be perceived differently at the announcement date, suggesting that the overall positive CARs cannot be attributed to leverage. As in model (1) we do not find a significant relationship for the common shareholder with shareholdings at 5%. When looking at shareholdings of 10% or more in bidder and/or target, we find that the coefficient of the large common shareholder is positive and significant at the 0.1 level. These results imply that bidder’s CARs for mergers with a large common shareholder are 2.3% greater than for mergers without a common shareholder. In models (5) and (6) we examine the relation between the CAR and the common shareholder’s ownership percentages in the bidder and target separately. For the ownership in the bidder company we find a coefficient of 0.063, which is significant at a 0.1 level. Evaluating this at the mean suggests that an increase in the ownership by one standard deviation is associated with a 1.11% increase in the average CAR. The common shareholder’s ownership in the target company also has a significant positive relation with CAR at a 0.05 level. Evaluating the coefficient, we find that an increase in the ownership by one standard deviation is associated with a 1.54% increase in the average CAR. These results confirm that the common shareholder’s ownership relates positively to CAR and that ownership in the target has a more positive influence than ownership in the bidder company. In model (5) we also find a significant coefficient of -0.034 for rescue mergers, confirming our earlier findings of the negative influence on abnormal return. Overall we find that CARs do not vary according affiliation with a main bank and the level of leverage in our regression results, which is in contrast to Kang et al. (2000). The regression shows that the existence of a common shareholder has an important influence on CAR.

185

197 Chapter 9 Merger activity

Regression analysis: Volume Level In table 9-6, the regression results on excess volume level are shown in models (7) and (8). We only report the Excess Volume-variant of the CAR test models (1) and (2) because the other tests show similar results. Model (7) confirms the results from the univariate analysis. For all companies the excess volume is positively influenced by the relative size of the transaction at a significance level of 0.05. Being a member of a horizontal keiretsu results in a lower trading volume than an independent company. This is an interesting finding because it suggests that the cross- shareholding structures have not been replaced yet. In our other tests, however, we have seen that it does not have a positive or negative effect on the stock price at the announcement date. In model (8) we examine the trading volume of bidder companies affiliated with a main bank, based on outstanding main bank loan percentage. We get the same results as model (7) with reference to the transaction size and membership of a horizontal keiretsu. For these bidder companies, use of main bank loans and a merger with a listed target company result in a lower trading volume.

9-3-3 Conclusions M&A activity in Japan has been low relative to other countries, which was attributed to the influence of industrial groups and the main bank system. Main banks as well as group membership guaranteed a certain level of corporate monitoring and stable and cross- shareholdings, which substituted the market for corporate control. After the collapse of the bubble in the 1990s, the subsequent economic recession, and major legislative and accounting changes, Japanese companies were forced to question the efficiency of the traditional Japanese system of corporate governance. Consequently, Japanese companies started to engage in restructuring and M&A activities. This study supports the hypothesis that Japanese corporate control has evolved over the past decade. The presence of main banks or membership of a keiretsu industrial group do not appear to affect the market’s reaction to the announcement of a merger, in contrast to studies done over earlier time periods. On average, bidders show a positive abnormal return around the announcement date of about 1.4%. The largest abnormal return is realized in the 2 days before the announcement but gains are quickly lost thereafter. Interestingly, trading volume appears to increase only after the announcement is made. Insiders appear to be able to profit from superior knowledge on the imminent merger announcement, while the rest of the market is too late. Factors other than affiliation with a main bank or keiretsu membership do appear to have a positive association with announcement returns, such as the presence of a common shareholder holding shares in both the bidder and the target company and the period in which the merger took place (in/after 1998). For the volume effect we find an association with the size of the target company relative to the bidder company.

186

198 Section 9-4 Shareholder wealth of bidder and target companies

9-4 Shareholder wealth of bidder and target companies R When the main bank is the same for the bidder and target company, its involvement does not create shareholder wealth but is motivated to protect its own interests as creditor: same main bank mergers are not between two weak companies, but at least one of the merging companies is financially strong. R Bidder and target companies accrue positive pre-announcement returns, with bidder companies capturing approximately half the gains that accrue to target companies. R We find differential shareholder wealth effects in the bubble period (1982-1989), the early 1990s (1990-1996) and the late 1990s (1997-2003).

In chapter 2 and section 9-1 we explained that the second merger wave occurred after the stock and land price bubbles burst. In the preceding high-growth period companies had made a lot of investments, and these companies faced financial difficulties in the 1990s. A main bank, being interested in payment of interest and repayment of loans, will not be concerned in a period of high economic growth, but this attitude will change in times of economic recession. In this latter situation, companies with falling profitability or in financial distress are likely to be restructured by the main bank. In this section we examine whether rescue mergers occurred in Japan and how the strong relationship of a main bank with the merging companies influences merger activity. Morck and Nakamura (1999) show that main banks act in the interest of creditors in stead of shareholders. Bankers are appointed to the board of directors of companies with poor performance and low liquidity. Their evidence supports the “bank power hypothesis”. When we relate the bank power hypothesis to mergers, we can assume that the main bank of a company in financial distress will attempt to arrange a merger of a weak company with a stronger company. The merger will protect the main bank’s interests as a creditor and will not necessarily be intended to increase shareholder wealth. In contrast, Kang et al. (2000) find significant positive abnormal returns for bidder companies that are affiliated with a main bank; they conclude that the main bank enhances shareholder wealth. In previous research and in our own research in section 9-3, the influence of the main bank is examined by focusing on the bidder company. Kang et al. (2000) only mention the influence of the target company’s main bank with reference to the fact that the bidder is unlikely to overpay for the target as it can take advantage of the information of the main bank (p.2209). As we did not find significant results in tests on performance and the main bank in our research in section 9-3, we argue that the main bank of the target company should be most important in the analysis of the bank power hypothesis. When failing companies are involved in mergers, these are most likely the target companies in the transaction. The main bank of the failing company is therefore most concerned about its interests, most active in finding a way to secure these, and its role can only be deduced from the actions towards the failing company. We further argue that only when the merging companies have the same main bank, it is possible to maintain that the main bank engages in activities to protect its own interests. The same main bank will gain by arranging a merger of a company in financial distress with a financially strong company, as it can herewith secure its interests. Also, a same main bank that is concerned with its own interests will not coordinate mergers between two failing companies. In summary, if the main bank has a role in mergers in Japan, we can only find this in periods of economic recession and when the merging companies have the same main bank.

187

199 Chapter 9 Merger activity

This section reviews the results of section 9-3 and of Kang et al. (2000) by investigating 91 mergers between listed companies in the period 1982-2003. We examine whether the main bank enhances shareholder wealth (Kang et al. 2000) or acts in its own interest as creditor (Morck and Nakamura 1999). In our analysis we pay special attention to mergers in which a failing company is involved and to the specific macro-economic conditions, such as the business cycle and changes in the regulatory structure, as they can influence the type of mergers that companies engage in. In particular, we examine the following: First, we look into the sub-periods: (i) the period 1982-1989 (“1980s”), (ii) the period 1990-1996 (“Early 90s”), and (iii) the period 1997-2003 (“Late 90s”). The 1980s is characterized by a growing economy at an average annual GNP growth of 5.5% in which companies made a lot of investments in machinery and real estate. The investments led to increasing prices in the stock market and of land prices in the late 1980s. The stock market bubble burst at the end of 1989 and the land prices bubble in 1991. During the second period, Early 90s, the Japanese economic growth slowed down considerably and GNP- growth dropped to an annual average of 1.5%. A lot of companies faced financial difficulties and a lot of banks coped with non-performing loans. The third period, Late 90s, is typified by the aggravation of the financial difficulties and various amendments to laws related to M&A and ownership, such as the lifting of the ban on pure holding companies. In 1997 the severity of the non-performing loan problems of Japanese financial institutions became evident with bankruptcies of some large financial institutions. Second, we investigate the influence of the main bank. Previous research argues that the main bank relationships have started to weaken in recent years. The length of our sample period allows a good examination if and how the main bank’s influence changed in recent years. In particular, we focus on the influence of the main bank when it is the same for the merging companies. Third, we investigate the influence of profitability and liquidity on shareholder wealth and on the role of the main bank.

9-4-1 Data We look into domestic mergers between non-financial companies in the 22-year period from 1 January 1982 until 31 December 2003. Our sample consists of bidder and target companies listed on either the First Section or Second Section of the Tokyo Stock Exchange (TSE). To obtain our data-set we first collect information on all companies that were delisted from TSE during the sample period. Next, we investigate whether the delisted companies were engaged in a merger by examining all press articles related to mergers in the period 1982 to 2003. The press articles are from the Nihon Keizai Shimbun (Japan Economic Journal), Nikkei Sangyǀ Shimbun (Industrial Journal), Nikkei Ryǌtǌ Shimbun (Distribution Journal) and Nikkei Kinyǌ Shimbun (Finance Journal). If the company was engaged in a merger, we collect the initial public Announcement Date (AD) and the Effective Date (ED) of the merger from the press articles. The AD is defined as the day that the merger announcement appears in the press for the first time. We next look for the accounting data and stock price data of the merging companies. Our final sample contains 91 separate mergers over the period 1982-2003. Figure 9-4 shows the distribution of mergers over the sampling period. Two characteristics are evident, half of all merger transactions occur between 1998 and 2003 and the mergers occur counter-cyclical to the stock market. This appears to indicate that in Japan, rather

188

200 Section 9-4 Shareholder wealth of bidder and target companies

than shareholders, fixed claimants such as creditors and employees are most important in mergers.

Figure 9-4 Merger activity and Nikkei Index in the period 1981-2003

14 45,000

40,000 12

35,000 10 30,000

8 25,000 Index Cases 6 20,000

15,000 4 10,000

2 5,000

0 0

6 7 8 9 0 1 2 3 4 5 1 2 3 8 8 8 8 0 0 0 99 99 99 99 99 99 996 997 1982 1983 1984 1985 19 19 19 19 1 1 1 1 1 1 1 1 1998 1999 2000 20 20 20 Merger cases Nikkei Index Notes On the left vertical axis the number of mergers is given and on the right vertical axis the Nikkei Index is plotted, measured at year-end value. The sample consists of 91 mergers between Japanese bidder and target companies listed on the Tokyo Stock Exchange for which the announcement date of the merger is between 1 January 1982 and 31 December 2003. Information on all companies that were delisted from TSE during the sample period was collected. Next, it was investigated whether the delisted companies were engaged in a merger by investigating all press articles related to mergers in the period 1982 to 2003.

As explained above, we pay special attention to three sub-periods in the sample period. Table 9-7 shows that 19 mergers of the total sample occur in the 1980s, 21 mergers in the early 90s, and 51 mergers in the late 90s. We examine the main bank defined as a bank that is the company’s most important lender and belongs to the company’s largest 5 shareholders, as indicated in the publication keiretsu no kenkyǌ for the year of the announcement. Table 9-7 shows that target companies with a main bank amount to 67% of all mergers, and bidder companies with a main bank to 65%. In 31% of the merger cases the target and bidder company have the same main bank. Mergers in which the merging companies have a main bank are concentrated in the early 90s with 95% for bidder companies and 76% for target companies. In this period more than half of all mergers is between companies with the same main bank. The financial difficulties after the bubble period appear to have stimulated main banks to arrange mergers between related companies. In the periods 1980s and Late 90s the percentage of target companies with a main bank slightly exceeds bidder companies. Mergers involving companies with the same main bank are at 29% in the 1980s and 22% in the Late 90s. We define financial distress following Hoshi, Kashap and Scharfstein (1990) and select companies that experience a cash-flow crisis. We examine whether the company’s interest expense exceeds its operating income in (i) the last fiscal year prior to the merger or (ii) in

189

201 Chapter 9 Merger activity

two of the four years before the merger. Mergers involving a distressed bidder are most frequent in the 1980s with 58%. Mergers with a distressed target are at the same level in the 1980s and Early 90s with respectively 63% and 67% of the cases. Mergers involving two distressed companies are most frequent in the 1980s at 47%. Overall, one third of the mergers involve a distressed bidder and in half of all merger cases a distressed target is involved. The percentage of mergers involving two distressed companies is highest in the 1980s and lowest in the late 90s. Based on the Nikkei Needs industry classification we investigate whether the mergers are intra- or inter-industry mergers. In the 1980s most inter-industry mergers take place with 37% (seven out of the 19 merger cases). The percentage decreased to 24% in Early 90s and 10% in late 90s. This indicates that whereas the mergers in the 1980s were motivated by product extension or speculation, the 1990s are primarily aimed at increasing efficiency and cutting costs.

Table 9-7 Characteristics by period All 1980s Early 90s Late 90s Number of mergers 91 19 21 51 Bidder main bank 65% 53% 95% 57% Target main bank 67% 58% 76% 67% Same main bank 31% 29% 52% 22% Bidder distressed 35% 58% 33% 27% Target distressed 51% 63% 67% 39% Both distressed 27% 47% 29% 20% Inter-industry 19% 37% 24% 10% Notes The sample consists of 91 mergers between Japanese bidder and target companies listed on the Tokyo Stock Exchange for which the announcement date of the merger is between 1 January 1982 and 31 December 2003. Information on all companies that were delisted from TSE during the sample period was collected. Next, it was investigated whether the delisted companies were engaged in a merger by investigating all press articles related to mergers in the period 1982 to 2003. A main bank is defined as a bank that is a company’s most important lender and belongs to the company’s largest 5 shareholders, as indicated in keiretsu no kenkyǌ for the year of the announcement. A company is qualified as being in financial distress when interest expense exceeds its operating income in the last fiscal year prior to the merger, or in two of the four years prior to the merger. Based on the listing on the TSE we determine whether it is an inter-industry or intra-industry merger.

Table 9-8 shows descriptive statistics of the companies in the sample. Target companies have assets at a mean of 151 billion yen and median of 51 billion yen. The mean and median of bidders’ total assets are considerably higher at 690 billion yen and 199 billion yen. In confirmation of our findings in chapter 2, the merging companies were largest in size during the Early 90s. The means of total assets are 1,079 billion yen for the bidder companies and 208 billion yen for target companies. The size of the target company relative to the bidder company, on the other hand, is highest in the Late 1990s with a mean of 41.7%. This confirms our findings on intra-industry mergers in this period as being aimed at cutting costs. The diversification mergers in the 1980s, on the other hand, were predominantly with small targets. For target and bidder companies with a main bank we measure leverage as main bank loans divided by total debt and find means at respectively 7.3% and 4.5%. These main bank loan ratios are for both target companies and bidder companies highest in the 1980s. The increasing stock market and growing economy resulted in a lot of corporate investments partly financed by loans from the main bank. The ratio of bidder companies exceeds that of target companies in the 1980s, but in the other periods the ratio of targets is more than double that of bidder companies, increasing their bankruptcy costs. In Late 90s

190

202 Section 9-4 Shareholder wealth of bidder and target companies

the target companies’ mean is 8% compared to 3.1% for bidder companies. The main bank shareholdings are also highest in the 1980s with 6.0% in target companies and 5.3% in bidder companies. In the 1990s the ownership percentages of main banks are between 4% and 5% for the merging companies. Regarding the influence of common financial institutions, target companies have the highest shareholding ratio of 11.8% in the Early 90s, and bidder companies in the 1980s at 12.9%. We can see a strong decline in the Late 90s when the unwinding of stable shareholdings started. Mergers in which the bidder owns shares of the target fall from 63% in the 1980s to 40% in the Late 90s. Of the 41 mergers in the total sample, bidder companies have ownership of target companies at an average percentage of 31.1%. This ratio is stable over the entire sample period. The percentage of merger cases involving a common corporate shareholder increases from 21% in the 1980s to 37% in the Late 90s. The average ownership percentage of the common corporate shareholder was 30.1% for targets and 27.1% for bidders in the Late 90s. A large corporate shareholder is a large shareholder with ownership exceeding 10% in only one of the merging companies. A large corporate shareholder in a target company does not belong to the top ten shareholders of a bidder company, and vice versa. Target companies have a large corporate shareholder in 27 merger cases at an average shareholding of 22.8%, and bidder companies in 25 mergers at 26.6%. In the 1980s both the number of mergers involving a corporate shareholder and the ownership percentage is lowest.

191

203 Chapter 9 Merger activity

Table 9-8 Descriptive statistics of sample (in million yen) Target Bidder Variable Mean Median Mean Median Total Assets All 91 151,609 51,448 690,302 199,709 1980s 19 141,863 15,895 473,691 122,176 Early 90s 21 208,165 114,680 1,078,900 701,222 Late 90s 51 131,952 55,933 610,989 152,572 TotalAssetsTarget/TotalAssetsBidder All 91 0.453 0.343 1980s 19 0.396 0.138 Early 90s 21 0.412 0.278 Late 90s 51 0.492 0.417 TotalDebt/TotalAssets All 91 0.705 0.703 0.679 0.711 1980s 19 0.734 0.758 0.690 0.704 Early 90s 21 0.713 0.710 0.724 0.771 Late 90s 51 0.691 0.698 0.657 0.707 Main Bank Loans / Debt All 63 0.073 0.030 59 0.045 0.011 1980s 11 0.088 0.054 10 0.131 0.042 Early 90s 16 0.047 0.028 20 0.022 0.006 Late 90s 36 0.080 0.026 29 0.031 0.010 Main Bank Shareholding (ownership %) All 634.34.7594.34.5 1980s 11 6.0 4.8 10 5.3 5.1 Early 90s 16 4.5 4.8 20 3.8 4.0 Late 90s 36 3.7 4.3 29 4.2 4.5 Common Financial Institution Shareholders' loans / Debt All 69 0.078 0.047 0.064 0.042 1980s 15 0.036 0.029 0.048 0.039 Early 90s 20 0.071 0.062 0.086 0.044 Late 90s 34 0.101 0.045 0.058 0.038 Common Financial Institution Shareholders (Ownership %) All 69 9.8 7.8 10.7 10.0 1980s 15 11.6 8.9 12.9 13.9 Early 90s 20 11.8 10.3 11.8 11.7 Late 90s 34 7.8 7.1 9.0 7.8 Bidder ownership of Target (Ownership %) All 41 31.1 32.7 1980s 12 31.8 32.7 Early 90s 9 28.2 28.4 Late 90s 20 31.9 31.4 Common Corporate Shareholder (Ownership %) All 28 24.9 24.9 24.0 21.5 1980s 4 17.4 20.1 19.0 11.9 Early 90s 5 11.1 4.8 16.6 3.5 Late 90s 19 30.1 28.6 27.1 22.5 Large Corporate Shareholder (Ownership %) All 27 22.8 18.6 25 26.6 22.7 1980s 4 19.8 17.8 4 21.9 19.6 Early 90s 7 19.4 18.2 3 30.4 28.4 Late 90s 16 25.0 18.9 18 27.0 22.3 Notes The sample consists of 91 mergers between Japanese bidder and target companies listed on the Tokyo Stock Exchange for which the announcement date of the merger is between 1 January 1982 and 31 December 2003. Information on all companies that were delisted from TSE during the sample period was collected. Next, it was investigated whether the delisted companies were engaged in a merger by investigating all press articles related to mergers in the period 1982 to 2003. Accounting data is derived

192

204 Section 9-4 Shareholder wealth of bidder and target companies

from the Nikkei Needs Tapes. Data on shareholders of the bidder and target companies is retrieved from the publication keiretsu no kenkyǌ. A main bank is defined as a bank that is a company’s most important lender and belongs to the company’s largest 5 shareholders, as indicated in keiretsu no kenkyǌ for the year of the announcement.

Table 9-9 shows the return on assets (“ROA”) of the bidder and target companies for the entire period and by sub-period. The ROA of all bidder companies and target companies falls in the three years prior to the merger. After the merger the new company shows an increase in its ROA in the third year. The table also shows the ROA for the three sub- periods and we find strong differences between the profitability of the merged company. In the 1980s the ROA of bidder companies is high and falling slightly before the merger. It increases from 6.5% in year -3 to 7.3% in the year -2, but then falls to 5.5% in the year prior to the merger. The ROA of the target companies declines strongly from 4.8% in the year -3 to 1.7% in the year -1, but remains positive. The ROA in the year following the merger is lower than that of the bidder companies. It increases the following year but thereafter it falls. The sample of 19 mergers in this period consists for 47% of mergers between companies that are both in financial distress and for 29% of mergers in which both have the same main bank. The merger does not result in higher profitability after the merger for the new company. In the Early 1990s the bidder companies’ ROA falls slightly from 3.5% in year -3 to 2.8% in the year -1. The ROA of target companies is very low and turns negative in the year prior to the merger at -0.06%. The new company shows a ROA of 2.2% that increases slightly the following years. In the early 1990s we see that bidder companies have a reasonable profitability but that target companies’ profitability declines and turns negative in the year prior to the merger. In 52% of the mergers the companies have the same main bank, in 29% both companies are in financial distress and in 67% of the mergers the target company is in financial distress. The merger results in lower profitability for bidder companies in the year of the merger, but profitability increases thereafter. In the Late 1990s, the ROA of the bidder companies remains relatively stable and target companies show the highest ROA in the year prior to the merger (highest of the three years prior to the merger and compared to the preceding two periods). This leads the new company to have a ROA of 3.3% in the year of the merger, similar to the bidder companies’ ROA in the year before the merger. The following two years the ROA is at a lower level of 2.7%, but it increases to 3.8% in the third year after the merger. In the late 1990s both the target and bidder companies have good profitability and the profitability after the merger increases the third year. Mergers involving companies with the same main bank or between companies in financial distress are low at about 20% of all mergers.

193

205 Chapter 9 Merger activity

Table 9-9 ROA data of bidders and targets ROA n -3 -2 -1 0 1 2 3 All Bidder 91 Mean 0.042 0.042 0.036 0.032 0.030 0.029 0.034 Median 0.035 0.034 0.032 0.028 0.030 0.028 0.035 Target Mean 0.024 0.018 0.015 Median 0.026 0.020 0.017

1980s Bidder 19 Mean 0.065 0.073 0.055 0.039 0.042 0.036 0.034 Median 0.054 0.055 0.052 0.046 0.044 0.053 0.045 Target Mean 0.048 0.035 0.017 Median 0.049 0.030 0.013

Early 90s Bidder 21 Mean 0.035 0.032 0.028 0.022 0.024 0.026 0.027 Median 0.036 0.034 0.032 0.024 0.030 0.023 0.027 Target Mean 0.012 0.004 -0.006 Median 0.026 0.012 0.010

Late 90s Bidder 51 Mean 0.036 0.035 0.033 0.033 0.027 0.027 0.038 Median 0.031 0.031 0.026 0.028 0.020 0.022 0.036 Target Mean 0.020 0.018 0.024 Median 0.023 0.019 0.022 Notes The sample consists of 91 mergers between Japanese bidder and target companies listed on the Tokyo Stock Exchange for which the announcement date of the merger is between 1 January 1982 and 31 December 2003. Information on all companies that were delisted from TSE during the sample period was collected. Next, it was investigated whether the delisted companies were engaged in a merger by investigating all press articles related to mergers in the period 1982 to 2003. ROA is calculated as operating profit divided by total assets. The accounting data is retrieved from the Nikkei Needs Tapes.

9-4-2 Results Abnormal returns are computed using the event-study methodology, following Dodd and Warner (1983). Our test period starts 50 days prior to the announcement date and stops 50 days after. We calculate abnormal returns in three ways. First, similar to Section 9-3, compared to normal returns that are determined by calculating betas using daily return data over the 200 trading days preceding the test period, using the TOPIX index as our benchmark. Second, similar as the first calculation but the betas are at a weighted average. Third, we calculate abnormal returns as raw returns. We get similar results for each method of calculation and the calculations below are based on the third method. The daily abnormal return is compounded over various time intervals to get the cumulative average abnormal return (CAR). We use standard t-statistics to test the hypothesis that the average CARs are equal to zero. Table 9-10 shows several windows of abnormal price returns around the announcement date.

194

206 Section 9-4 Shareholder wealth of bidder and target companies

Table 9-10 CARs for Japanese targets and bidders in the period 1982-2003 All Companies (n = 91) CAR Main Bank CAR Not Main Bank (n = 28) (n = 63) Target CAR Bidder CAR Bidder Target Bidder Target Mean Median Mean Median Mean Mean Mean Mean [-50, 0] 0.109 0.098 0.039 0.014 -0.038 0.166 0.073 0.083 (0.000) (0.001) (0.034) (0.098) (0.170) (0.005) (0.001) (0.014) [-5, 0] 0.047 0.031 0.018 0.016 0.004 0.089 0.024 0.028 (0.000) (0.001) (0.023) (0.044) (0.789) (0.002) (0.011) (0.048) [-2, 0] 0.024 0.008 0.011 0.006 0.002 0.046 0.015 0.014 (0.031) (0.036) (0.074) (0.058) (0.849) (0.032) (0.041) (0.280) [-1, 0] 0.014 0.002 0.004 0.002 -0.012 0.041 0.011 0.002 (0.145) (0.115) (0.515) (0.554) (0.228) (0.020) (0.124) (0.834) [-1, +1] -0.010 0.000 -0.012 -0.012 -0.044 0.025 0.002 -0.026 (0.466) (0.446) (0.110) (0.087) (0.006) (0.296) (0.803) (0.131) [0, +2] -0.037 -0.041 -0.013 -0.011 -0.031 -0.015 -0.006 -0.046 (0.039) (0.010) (0.084) (0.033) (0.029) (0.639) (0.530) (0.031) [0, +5] -0.064 -0.066 -0.020 -0.022 -0.034 -0.051 -0.014 -0.069 (0.003) (0.001) (0.031) (0.020) (0.037) (0.144) (0.221) (0.009) [0, +50] -0.046 -0.031 -0.017 -0.019 -0.019 -0.079 -0.016 -0.031 (0.063) (0.058) (0.315) (0.194) (0.482) (0.095) (0.452) (0.281) [-2, +2] -0.020 -0.007 -0.006 -0.006 -0.023 0.012 0.002 -0.034 (0.260) (0.153) (0.503) (0.293) (0.138) (0.679) (0.845) (0.116) [-5, +5] -0.024 -0.034 -0.006 -0.007 -0.025 0.019 0.002 -0.043 (0.264) (0.286) (0.566) (0.432) (0.151) (0.612) (0.867) (0.098) [-50, +50] 0.056 0.081 0.019 -0.009 -0.051 0.068 0.049 0.050 (0.074) (0.122) (0.406) (0.786) (0.082) (0.250) (0.093) (0.175) Notes The sample consists of 91 mergers between Japanese bidder and target companies listed on the Tokyo Stock Exchange for which the announcement date of the merger is between 1 January 1982 and 31 December 2003. Information on all companies that were delisted from TSE during the sample period was collected. Next, it was investigated whether the delisted companies were engaged in a merger by investigating all press articles related to mergers in the period 1982 to 2003. The initial announcement date, i.e. the first day on which the information related to the announcement was public before the end of the trading day, is defined as the day that the merger announcement appears in the press for the first time. The press articles from the Nihon Keizai Shimbun (Japan Economic Journal), Nikkei Sangyǀ Shimbun (Industrial Journal), Nikkei Ryǌtǌ Shimbun (Distribution Journal), and Nikkei Kinyǌ Shimbun (Finance Journal) are investigated.

Total Sample We find that the stock price of the target company starts to rise as early as 50 days before the announcement. The total increase (adjusted for the market) from day –50 to day 0 is 10.9%, of which approximately half occurs in the five days preceding the announcement of the merger (4.7%). By the end of day +5, the gain for target companies is down to 4.5% (cumulative gain from day –50 to day +5). By the end of day +50, target shares recover somewhat to show an average cumulative gain of 5.6%. In contrast to the U.S. evidence, we find that bidder companies in Japan enjoy positive gains in the period leading up to the announcement of the merger. The bidder companies’ CAR from day –50 to day 0 is 3.9%. Immediately after the merger announcement, the bidder share price falls (as was the case with the target). The cumulative return from day –50 to day +5 is 1.9% for the bidder, identical to that of the target company during the same interval. There appears to be no recovery in bidder share prices in the following days – the cumulative bidder return from day +5 to day +50 is insignificant. Based on the data of all companies, we are unable to explain the loss in target share value immediately following the merger announcement – cancelled mergers are very rare, in fact, non-existent in our sample. The separation of mergers by presence and absence of a same main bank provides us more information on the negative CARs at/after the announcement date.

195

207 Chapter 9 Merger activity

Regarding the CARs for merging companies that have the same main bank, we do not find any significant changes in the share price prior to the announcement for either the bidder or target company. At and after the announcement date, on the other hand, the CARs of both merging companies turn negative at a significant level. The involvement of the main bank has a strong negative effect on shareholder wealth. Over the period day 0 to day +5 the CAR for target companies is at -11% and for bidder companies at -3.4%. In contrast, the CARs of companies in mergers without the influence of a same bank show significant positive gains in the period leading up to the announcement of the merger. During the announcement period the gains are insignificantly different from zero, and the overall effect of the absence of a same main bank is positive for both bidder and target companies. Next we examine how characteristics of the merging companies influence the abnormal returns for the period [-1,+1] (table 9-11 and table 9-12). We measure total bank loan leverage as total loans divided by total assets and find that bidder companies with a ratio above the median have a mean CAR at -2.4% and a median CAR at -1.6%, both significant at the 0.05 level. These CARs are not significantly different from companies with leverage below the median, and the other CARs related to leverage are all insignificantly different from zero. Ownership of a bidder company in the target company, and by a common shareholder in the target company and the bidder company does not have an important influence over CARs. Presence of a large corporate shareholder and being member of the same keiretsu has a minor negative influence on the bidder’s abnormal returns; the median CAR of bidder companies with a large corporate shareholder is significantly negative and distinguishable from the median of companies that do not have a large corporate shareholder. Inter-industry mergers are significantly negative for target companies. The mean is negative at -6.8% and the median at -4.7%, both distinguishable from the insignificant CARs of target companies in intra-industry mergers.

196

208 Section 9-4 Shareholder wealth of bidder and target companies

Table 9-11 CAR [-1, +1] for Japanese bidders and targets categorized by characteristics of the merging companies Target CAR [-1,+1] Bidder CAR [-1,+1] Mean Median t-test Wilcoxon Mean Median t-test Wilcoxon All 91 -0.010 0.000 -0.012 -0.012 (0.466) (0.446) (0.110) (0.087) Total bank loan ratio above 46 -0.017 0.000 0.483 -0.024 -0.016 1.606 sample median (0.259) (0.368) (0.630) (0.035) (0.033) (0.112) Total bank loan ratio below 45 -0.003 0.000 0.373 0.000 0.000 1.353 sample median (0.888) (0.803) (0.709) (0.998) (0.804) (0.176) Bidder and target have common 28 0.010 0.003 0.971 -0.007 0.001 0.536 shareholder (0.691) (0.747) (0.334) (0.637) (0.542) (0.595) Bidder and target do not have 63 -0.019 -0.001 0.989 -0.014 -0.013 0.557 common shareholder (0.255) (0.249) (0.323) (0.101) (0.077) (0.577) Bidder with ownership of more 26 -0.021 -0.001 0.480 -0.011 0.007 0.126 than 20% in target (0.430) (0.438) (0.632) (0.474) (0.979) (0.900) Bidder with no or lower than 20% 65 -0.006 0.000 0.422 -0.013 -0.017 1.208 ownership in target (0.721) (0.669) (0.673) (0.154) (0.058) (0.227) Bidder and target have csh over 16 0.006 0.009 0.534 0.010 0.012 1.359 20% (0.819) (0.755) (0.594) (0.481) (0.423) (0.178) Bidder and target do not have csh 75 -0.014 0.000 0.751 -0.017 -0.014 1.632 over 20% (0.397) (0.340) (0.453) (0.054) (0.021) (0.103) Bidder and csh with ownership of 44 -0.010 0.000 0.440 -0.005 0.007 0.948 more than 20% in target (0.642) (0.577) (0.661) (0.695) (0.960) (0.346) Bidder and csh with no or lower 47 -0.010 0.000 0.359 -0.019 -0.018 1.751 than 20% ownership in target (0.575) (0.588) (0.719) (0.044) (0.012) (0.080) Big corporate shareholder in 27 0.021 0.000 1.481 -0.017 -0.022 0.404 target or bidder (0.241) (0.294) (0.142) (0.324) (0.047) (0.688) Not big corporate shareholder in 64 -0.023 -0.001 1.499 -0.010 0.002 1.738 target or bidder (0.200) (0.134) (0.134) (0.216) (0.549) (0.082) Member of same keiretsu 47 -0.026 0.000 1.196 -0.018 -0.015 0.816 (0.156) (0.296) (0.235) (0.085) (0.214) (0.417) Not member of same keiretsu 44 0.007 0.000 0.647 -0.006 -0.011 0.171 (0.743) (0.940) (0.517) (0.604) (0.259) (0.864) Intra-industry merger 74 0.003 0.001 2.035 -0.011 -0.011 0.308 (0.842) (0.739) (0.045) (0.204) (0.156) (0.759) Inter-industry merger 17 -0.068 -0.047 2.449 -0.017 -0.018 0.270 (0.018) (0.011) (0.014) (0.283) (0.394) (0.787) Notes The sample consists of 91 mergers between Japanese bidder and target companies listed on the Tokyo Stock Exchange for which the announcement date of the merger is between 1 January 1982 and 31 December 2003. Information on all companies that were delisted from TSE during the sample period was collected. Next, it was investigated whether the delisted companies were engaged in a merger by investigating all press articles related to mergers in the period 1982 to 2003. The initial announcement date, i.e. the first day on which the information related to the announcement was public before the end of the trading day, is defined as the day that the merger announcement appears in the press for the first time. The press articles from the Nihon Keizai Shimbun (Japan Economic Journal), Nikkei Sangyǀ Shimbun (Industrial Journal), Nikkei Ryǌtǌ Shimbun (Distribution Journal), and Nikkei Kinyǌ Shimbun (Finance Journal) are investigated. A main bank is defined as a bank that is a company’s most important lender and belongs to the company’s largest 5 shareholders, as indicated in keiretsu no kenkyǌ for the year of the announcement. A company is qualified as being in financial distress when interest expense exceeds its operating income in the last fiscal year prior to the merger, or in two of the four years prior to the merger.

Main Bank Regarding the influence of the companies’ main bank(s) and financial distress on abnormal return of the merging companies, we first look into the influence of a target with a main bank on the CARs of target and bidder companies. Subsequently we investigate the influence of a bidder with a main bank, and finally we relate these to the effect on CARs when the merging companies have the same main bank. For both bidder and target companies, the return in same main bank mergers is more negative than in mergers where

197

209 Chapter 9 Merger activity

the main bank is different, or only one of the merging companies has a main bank relationship.

Targets with a main bank - We find that mergers involving target companies with a main bank result in a negative mean and median CAR for target companies at respectively - 2.8% and -1.0%, with the latter being significant at a 0.1 level. This mean and median CAR is distinguishable from a target companies’ CAR in mergers involving targets without a main bank. Target companies’ CAR of mergers involving a target without a main bank have a mean at 2.6% and median at 0.5%, but both are not significantly different from zero. We find a significant negative CAR for bidder companies at a mean of -1.7% and median of -1.3% in mergers involving a target company with a main bank. These CARs are not distinguishable from bidder CARs in mergers involving targets without a main bank. Bidder returns in mergers with targets without a main bank are not significantly different from zero.

Table 9-12 CAR [-1, +1] for Japanese bidders and targets categorized by main bank and financial distress Target CAR [-1,+1] Bidder CAR [-1,+1] Mean Median t-test Wilcoxon Mean Median t-test Wilcoxon Target with main bank (mb) 61 -0.028 -0.010 1.844 -0.017 -0.013 0.942 (0.122) (0.068) (0.069) (0.094) (0.064) (0.349) Target without mb 30 0.026 0.005 2.014 -0.002 -0.005 0.663 (0.215) (0.187) (0.044) (0.842) (0.864) (0.507) Bidder with mb 59 -0.041 -0.010 3.243 -0.025 -0.016 2.517 (0.011) (0.018) (0.002) (0.005) (0.007) (0.014) Bidder without mb 32 0.048 0.017 2.876 0.013 0.009 1.999 (0.049) (0.047) (0.004) (0.330) (0.537) (0.046) Target & Bidder same mb 28 -0.068 -0.061 2.927 -0.043 -0.025 2.938 (0.016) (0.026) (0.004) (0.006) (0.010) (0.004) Target & Bidder not same mb 63 0.016 0.000 2.330 0.002 0.000 2.283 (0.301) (0.472) (0.020) (0.805) (0.886) (0.022) Target in distress 46 -0.018 -0.001 0.593 -0.030 -0.029 2.474 (0.299) (0.296) (0.555) (0.002) (0.001) (0.015) Target not in distress 45 -0.002 0.000 0.635 0.006 0.009 3.401 (0.934) (0.990) (0.525) (0.589) (0.245) (0.001) Bidder in distress 32 -0.023 -0.001 0.693 -0.024 -0.025 1.206 (0.309) (0.332) (0.490) (0.033) (0.060) (0.231) Bidder not in distress 59 -0.003 0.000 0.748 -0.005 0.000 1.425 (0.863) (0.890) (0.454) (0.583) (0.496) (0.154) Target and bidder in distress 25 -0.009 0.000 0.029 -0.030 -0.029 1.482 (0.729) (0.809) (0.977) (0.016) (0.021) (0.142) Target and bidder not in distress 66 -0.010 0.000 0.049 -0.005 0.000 1.872 (0.524) (0.520) (0.961) (0.569) (0.591) (0.061) Notes The sample consists of 91 mergers between Japanese bidder and target companies listed on the Tokyo Stock Exchange for which the announcement date of the merger is between 1 January 1982 and 31 December 2003. Information on all companies that were delisted from TSE during the sample period was collected. Next, it was investigated whether the delisted companies were engaged in a merger by investigating all press articles related to mergers in the period 1982 to 2003. The initial announcement date, i.e. the first day on which the information related to the announcement was public before the end of the trading day, is defined as the day that the merger announcement appears in the press for the first time. The press articles from the Nihon Keizai Shimbun (Japan Economic Journal), Nikkei Sangyǀ Shimbun (Industrial Journal), Nikkei Ryǌtǌ Shimbun (Distribution Journal), and Nikkei Kinyǌ Shimbun (Finance Journal) are investigated. A main bank is defined as a bank that is a company’s most important lender and belongs to the company’s largest 5 shareholders, as indicated in keiretsu no kenkyǌ for the year of the announcement. A company is qualified as being in financial distress when interest expense exceeds its operating income in the last fiscal year prior to the merger, or in two of the four years prior to the merger.

198

210 Section 9-4 Shareholder wealth of bidder and target companies

Bidderswithamainbank - The target companies’ CAR in mergers involving bidder companies with a main bank is significantly negative at a 0.05 level, with a mean of -4.1% and a median of -1.0%. This CAR is distinguishable from the CAR involving bidder companies without a main bank at a 0.01 level. Target companies’ CAR in mergers with bidder companies without a main bank is significant at 4.8% (mean) and 1.7% (median). Bidder companies with a main bank have a significant negative CAR at a mean and median of respectively -2.5% and -1.6%. The CAR is significantly different from the CAR of bidder companies without a main bank and the latter is not significant.

Same main bank – In mergers where target companies and bidder companies have the same main bank, the CAR of target companies is significantly negative at a mean of -6.8% and median of -6.1%. These returns are also significantly different from returns of target companies in mergers in which the bidder company and target company do not have the same main bank. The mean and median of a target companies’ CAR in this latter group is not significantly different from zero. Bidder companies have a mean and median of respectively -4.3% and -2.5% in same main bank mergers. These returns are significantly different from those of mergers in which the merging companies do not have the same main bank.

Financial distress Table 9-12 shows the results of our tests to examine the influence of financial distress of the merging companies as well. Similar to the main bank, we look into the influence of financial distress of target and bidder companies individually and when both are in financial distress. Our discussion of the results is slightly different as we describe the effect of the three types of financial distress on CAR. We first look into the CAR of target companies and then turn to the CAR of bidder companies. Target CAR - The table shows that the financial condition of the target and/or the bidder company does not have any significant influence over the CARs of target companies. The CAR for target companies is not significantly different from zero when target companies are in financial distress, or when they are not in financial distress. A similar pattern is visible in case the bidder company is in financial distress, or both companies are in financial distress. Bidder CAR - The abnormal returns of bidder companies are significantly negative for all merger cases in which a company in financial distress is involved. In mergers in which the target or both companies is/are in financial distress, the mean and median CAR is - 3.0% and -2.9% respectively. In merger cases involving a target in financial distress, the means and the medians of the returns are distinguishable from mergers with targets not in financial distress at a 0.05 level. In case the bidder company itself is in financial distress, the mean is -2.4% and the median is -2.5%. Mergers not involving companies in financial distress result in means and medians of bidder returns that are not significantly different from zero.

199

211 Chapter 9 Merger activity

Additional tests Above we described that the CAR of the target and bidder companies is, in particular, negatively influenced in mergers where the merging companies have the same main bank. A significant negative impact on the returns of bidder companies is also found in mergers involving a target in financial distress. To examine these results in more detail, we conduct various tests for groups of target and bidder companies according to the following two characteristics: (i) merger cases in which the merging companies have the same main bank and (ii) merger cases with a target in financial distress. For these merger cases we investigate what influences the abnormal returns of the target and bidder companies. We perform tests on the following characteristics, of which we discuss only those with significant results below: (i) target with a main bank, (ii) bidder with a main bank, (iii) merging companies with the same main bank, (iv) distressed target, (v) distressed bidder, (vi) both merging companies distressed, (vii) industry related merger, (viii) keiretsu related, (ix) total debt/total assets ratio, (x) corporate common shareholder, and (xi) sub-periods.

Same main bank : sub-sample In our tests for the 28 mergers involving merging companies with the same main bank, we only find significant differences between the three sub-periods. The results of our tests for the 1980s, Early 90s and Late 90s are shown in table 9-13. Before we discuss how the sub- periods influence the CARs of bidder and target companies with the same main bank, we look briefly into the total sample regarding the sub-periods. Panel A of Table 9-13 indicates that target companies’ CAR is not significantly different from zero in any of the sub-periods. The returns of the target companies in the periods are not distinguishable from each other as well. The bidder CAR has a mean of -3.1% and median of -3.9% in the Early 90s, which are significant at a 0.01 level. In the other two periods the bidder returns are not significantly different from zero. In the comparison of Early 90s with Late 90s, only the median is distinguishable at a 0.05 level. These results indicate that when we examine the total sample, we cannot find significant differences between the sub-periods. We will next turn to the influence of the same main bank in the sub-periods. Thereafter, we also look into the returns in the sub-periods for mergers between companies that do not have the same main bank.

200

212 Section 9-4 Shareholder wealth of bidder and target companies

Table 9-13 CAR [-1, +1] for Japanese bidders and targets by period and main bank Target CAR [-1,+1] Bidder CAR [-1,+1] Mean Median t-test Wilcoxon Mean Median t-test Wilcoxon Panel A: Examination by period 1980s 19 -0.046 -0.004 -0.016 0.003 (0.161) (0.207) (0.266) (0.457) Early 90s (1990-1996) 20 -0.022 0.000 -0.031 -0.039 (0.271) (0.486) (0.000) (0.001) Late 90s (1997-2003) 51 0.008 0.000 -0.003 0.000 (0.682) (0.796) (0.809) (0.921) Comparison 1980s - Early 90s 0.646 0.826 1.027 1.273 (0.522) (0.409) (0.311) (0.203) Comparison Early 90s- Late 90s 0.909 0.830 1.468 2.416 (0.367) (0.407) (0.147) (0.016) Comparison 1980s - Late 90s 1.426 1.347 0.604 0.806 (0.158) (0.178) (0.548) (0.421) PanelB:MainBank Bidder and target with same main bank 1980s 6 -0.051 -0.038 0.024 0.016 (0.604) (1.000) (0.155) (0.219) Early 90s (1990-1996) 11 -0.030 0.001 -0.027 -0.029 (0.306) (0.549) (0.008) (0.065) Late 90s (1997-2003) 11 -0.116 -0.082 -0.096 -0.122 (0.010) (0.022) (0.007) (0.065) Comparison 1980s - Early 90s 0.280 0.050 3.348 2.462 (0.784) (0.960) (0.004) (0.014) Comparison Early 90s- Late 90s 1.885 1.773 2.336 1.839 (0.074) (0.076) (0.030) (0.066) Comparison 1980s - Late 90s 0.779 0.452 2.971 2.563 (0.448) (0.651) (0.010) (0.010) Bidder and target not same main bank 1980s 13 -0.043 -0.004 -0.034 -0.029 (0.076) (0.126) (0.067) (0.108) Early 90s (1990-1996) 10 -0.014 0.000 -0.035 -0.044 (0.643) (0.554) (0.013) (0.033) Late 90s (1997-2003) 40 0.042 0.018 0.023 0.012 (0.045) (0.073) (0.024) (0.038) Comparison 1980s - Early 90s 0.802 0.930 0.047 0.217 (0.431) (0.352) (0.963) (0.828) Comparison Early 90s- Late 90s 1.297 1.285 2.845 2.923 (0.201) (0.199) (0.007) (0.004) Comparison 1980s - Late 90s 2.245 2.253 2.906 2.636 (0.029) (0.024) (0.005) (0.008) Notes The sample consists of 91 mergers between Japanese bidder and target companies listed on the Tokyo Stock Exchange for which the announcement date of the merger is between 1 January 1982 and 31 December 2003. Information on all companies that were delisted from TSE during the sample period was collected. Next, it was investigated whether the delisted companies were engaged in a merger by investigating all press articles related to mergers in the period 1982 to 2003. The initial announcement date, i.e. the first day on which the information related to the announcement was public before the end of the trading day, is defined as the day that the merger announcement appears in the press for the first time. The press articles from the Nihon Keizai Shimbun (Japan Economic Journal), Nikkei Sangyǀ Shimbun (Industrial Journal), Nikkei Ryǌtǌ Shimbun (Distribution Journal), and Nikkei Kinyǌ Shimbun (Finance Journal) are investigated. A main bank is defined as a bank that is a company’s most important lender and belongs to the company’s largest 5 shareholders, as indicated in keiretsu no kenkyǌ for the year of the announcement.

Sub-periods: same main bank - Target companies have negative returns in the periods 1980s and Early 90s but these are not significantly different from zero. In the Late 90s we find a significant negative mean of -11.6% and median of -8.2% at a 0.05 level. The

201

213 Chapter 9 Merger activity

returns between the Early 90s and the Late 90s are significantly distinguishable from each other. Bidder companies have positive but insignificant returns in the 1980s. The abnormal returns are significantly negative in the Early 90s and Late 90s, with means at respectively -2.7% and -9.6%. The comparisons of the bidders’ returns between the sub-periods are all significant. We can conclude that same main bank mergers resulted in significant negative returns for target companies in only the Late 90s. Returns of bidders in these mergers were significantly negative in Early 90s, significantly different from the 1980s, and deteriorated even more in the Late 90s. Same main bank mergers during the 1980s did not result in abnormal returns, these mergers resulted in negative abnormal returns for bidder companies during the Early 90s, and for both merging companies in the Late 90s. Sub-periods: not same main bank - When we turn to the companies in mergers without a same main bank, we see that the targets’ returns had a significant negative mean in the 1980s at -4.3%. In the Early 90s the returns are insignificant, but in the Late 90s the returns turn positive. In the Late 90s, the mean and median of target companies are respectively 4.2% and 1.8%, significant at a 0.1 level. The returns of the 1980s and Late 90s are significantly distinguishable from each other. For bidder companies we find a mean of - 3.4% for returns in the 1980s, significant at a 0.1 level. The returns in the Early 90s are significantly negative at a mean of -3.5% and median of -4.4%. In the Late 90s the returns turn positive and have an average of 2.3%. The Late 90s are distinguishable from the returns in the 1980s and the Early 90s at a 0.01 level. In contrast to same main bank mergers, we find for not same main bank mergers significant negative returns for the merging companies in the 1980s. In the Early 90s we find similar results for merging companies with and without the same main bank; insignificant returns for the target companies and significant negative returns for the bidder companies. This also explains that only in this period the total returns in Panel A of table 9- 13 are significant for bidder companies. In the Late 90s we find a significant positive abnormal return for not same main bank companies, an important difference with the significant negative abnormal returns for same main bank mergers. Overall, we find that in the 1980s mergers between companies with the same main bank resulted in insignificant positive returns for the bidder companies and insignificant negative returns for target companies. In mergers between companies without a same main bank, the merging companies had significant negative CARs, for target companies at - 4.3% and bidder companies at -3.4%. Table 9-14 indicates the financial health of the merging companies by period and presence of a same main bank. The table shows that in this period of economic growth, the main bank was primarily involved in mergers between strong companies. Mergers without a same main bank were for 62% between two weak companies. Most interesting in this period is that, whereas in not same main bank mergers two weak companies were combined as an attempt to survive, the same main bank did not engage in these activities as it would not secure its interests as creditor. In the Early 90s we find a negative abnormal return for bidder companies irrespective of being related to a main bank or not. The CARs of bidder companies in same main bank mergers are slightly less negative than in not same main bank mergers. The main difference is that in the main bank mergers, the mergers were predominantly (45%) between a strong bidder company and weak target company. With the merger, the same main bank tries to secure its own interests at the weak target company. The merging companies in not same main bank were both weak in 40% of the cases and, similar to the

202

214 Section 9-4 Shareholder wealth of bidder and target companies

1980s, appear to be an attempt to improve profitability by combining two failing companies.

Table 9-14 Classification of merging companies by financial health ALL Same Main Bank Not Same MB Strong Weak Strong Weak Strong Weak Bidder Bidder Bidder Bidder Bidder Bidder ALL Strong Target 42% 8% 43% 11% 41% 6% Weak Target 23% 27% 29% 18% 21% 32%

1980s Strong Target 26% 11% 67% 17% 8% 8% Weak Target 16% 47% 0% 17% 23% 62%

Early 90s Strong Target 29% 5% 27% 9% 30% 0% Weak Target 38% 29% 45% 18% 30% 40%

Late 90s Strong Target 53% 8% 45% 9% 55% 8% Weak Target 20% 20% 27% 18% 18% 20% Notes A strong company is not in financial distress, a weak company is in financial distress. We define companies as being in financial distress when interest expense exceeds operating income, the interest coverage ratio is lower than 1, in (i) the last fiscal year prior to the merger, or (ii) in two of the four years before the merger.

In the Late 90s the same main bank mergers have a strong negative effect on abnormal return of both the target and bidder companies. We see a strategy of the main bank that is slightly different from the Early 90s; mergers between a strong bidder and weak target at 27% and occur between two strong companies at 45%. Regarding the strong company mergers, in combination with the fact that they do not create shareholder wealth, we assume that the increasing amount of banks’ non-performing loans have encouraged the same main banks to arrange mergers of strong(er) companies to prevent more financial problems. Similar to the preceding periods, the main bank is not involved in mergers contrary to its own interests. In the not same main bank mergers we see that the fact that both merging companies are strong is reflected in the CARs of the target and bidder companies.

Target in financial distress: sub-sample For the 46 mergers involving a target company in financial distress, we only find significant differences for the factors (i) bidders with a main bank and (ii) intra- and inter- industry mergers.

203

215 Chapter 9 Merger activity

Bidders with main bank - Our sample consists of 29 bidder companies with a main bank that merge with a target company in financial distress78. The mean abnormal return of a distressed target company is significantly negative at -4.1% when the bidder company is affiliated with a main bank. In mergers with a bidder company not affiliated with a main bank the mean is not significantly different from zero. The means of the abnormal returns of target companies that merge with main bank bidders and non-main bank bidders, are distinguishable at a 0.1 level. The bidder companies with a main bank that merge with a target company in financial distress have a significantly negative mean abnormal return at -4.3%. The mean and median of the abnormal return of bidder companies without a main bank is not significantly different from zero. The means of the abnormal returns of main bank bidders and non-main bank bidders merging a target in financial distress are distinguishable at a 0.1 level.

Table 9-15 Target in financial distress Target CAR [-1,+1] Bidder CAR [-1,+1] Mean Median t-test Wilcoxon Mean Median t-test Wilcoxon Bidder with mb 29 -0.041 -0.010 1.734 -0.043 -0.032 1.911 (0.046) (0.442) (0.090) (0.000) (0.008) (0.063) Bidder without mb 17 0.020 0.009 1.593 -0.007 -0.018 1.274 (0.534) (0.804) (0.111) (0.676) (0.332) (0.203) Intra-industry 35 0.004 0.000 2.419 -0.029 -0.028 0.229 (0.826) (0.597) (0.020) (0.011) (0.017) (0.820) Inter-industry 11 -0.089 -0.065 2.369 -0.034 -0.043 0.155 (0.030) (0.065) (0.018) (0.118) (0.227) (0.877) Notes The sample consists of 91 mergers between Japanese bidder and target companies listed on the Tokyo Stock Exchange for which the announcement date of the merger is between 1 January 1982 and 31 December 2003. Information on all companies that were delisted from TSE during the sample period was collected. Next, it was investigated whether the delisted companies were engaged in a merger by investigating all press articles related to mergers in the period 1982 to 2003. The initial announcement date, i.e. the first day on which the information related to the announcement was public before the end of the trading day, is defined as the day that the merger announcement appears in the press for the first time. The press articles from the Nihon Keizai Shimbun (Japan Economic Journal), Nikkei Sangyǀ Shimbun (Industrial Journal), Nikkei Ryǌtǌ Shimbun (Distribution Journal), and Nikkei Kinyǌ Shimbun (Finance Journal) are investigated. A company is qualified as being in financial distress when interest expense exceeds its operating income in the last fiscal year prior to the merger, or in two of the four years prior to the merger. A main bank is defined as a bank that is a company’s most important lender and belongs to the company’s largest 5 shareholders, as indicated in keiretsu no kenkyǌ for the year of the announcement. Based on the listing on the TSE we determine whether it is an inter-industry or intra-industry merger.

Intra- and inter-industry mergers - Mergers involving targets in financial distress are intra- industry in 35 cases and inter-industry in 11. In inter-industry merger cases the returns of target companies are significantly negative at a mean of -8.9% and median of -6.5%. The cumulative abnormal returns in intra-industry mergers are not significantly different from zero, but the returns are distinguishable from inter-industry mergers at a 0.05 level. Bidder companies show a significantly negative return for intra-industry mergers with targets in financial distress at a mean of -2.9% and a median of -2.8%. Inter-industry mergers do not have a significant effect on bidders’ abnormal returns, and are not distinguishable from returns in intra-industry mergers.

78 The bidder companies with a main bank were distributed in the three sub-periods as follows: 1980s in 4 mergers (of 12), early 90s in 14 mergers (of 14), and in late 90s 11 mergers (of 20).

204

216 Section 9-5 Conclusion

9-4-3 Conclusions The market for corporate control in Japan behaves very different from that in the U.S. Using a sample of 91 mergers in the period 1982-2003, we document several distinctive features of this market in Japan. First, we show that in stark contrast to the pro- cyclical U.S. merger waves, mergers in Japan tend to be counter-cyclical, both with respect to the general economy as well as with respect to stock market valuations. Second, and again in contrast to the U.S. experience, we find that a significant fraction of Japanese mergers are orchestrated by the main banks; in such cases, a striking pattern emerges. When the main bank is the same for the bidder and target company, its involvement does not create shareholder wealth. The performance of at least one of the merging companies is strong, indicating that the same main bank is primarily motivated to protect its own interests as creditor. This was especially evident in the period after the stock price bubble burst, when the same main bank arranged mergers of weak borrowers with a financially strong buyer. In the entire period the same main bank’s involvement in mergers between two weak companies is low. Other distinctive features of Japanese mergers are the positive pre-announcement returns accruing to both bidders and targets, with bidders capturing approximately half the gains that accrue to target companies. The involvement of the same main bank has a strong negative effect on shareholder wealth at the announcement date, whereas in not same main bank mergers the merging companies show significant gains in the period leading up to the announcement. We also find differential shareholder wealth effects in the bubble period (1982-1989), the early 1990s, and the post-financial regulation regime (1997- 2003). Overall, our results point to a market for corporate control that is distinctly less shareholder-centred than that in the U.S. and one where creditors play an important, perhaps dominant, role.

9-5 Conclusion In the first section of this chapter we related the M&A activity, as discussed in chapter 2, to merger motives and drivers for merger wave. In the pre-war period we did not find merger waves. Mergers were motivated by synergy in the electricity sector and the adaptive motive appears to be most important in other sectors. A lot of companies fell in financial distress during the 1920s and the mergers appear to have been aimed at rescuing failing companies. In the 1960s, during the first merger wave, mergers were aimed at preventing excessive competition and achieving economies of scale in order to be able to compete with foreign companies (synergy motive). The first period of the second wave is dominated by mergers between companies in trouble due to the economic business cycle (adaptive motive). In the second period of the scond merger wave we find that companies also engaged in mergers for strategic purposes. With previous research we show that the main bank plays an important role in monitoring companies. The strong relationship between the main bank and the company allows the bank to intervene quickly when the company faces financial difficulty. Previous research shows considerable positive abnormal returns for U.S. target companies at the announcement date, but for Japanese target companies negative results (or only slightly positive) are found. In our own empirical research of sections 9-3 and 9-4 we investigate the role of the main bank in mergers with two event studies. In our first study we investigate whether the main bank increases shareholder wealth of bidder companies as a result of its knowledge and

205

217 Chapter 9 Merger activity

good relations with the company. We do not find significant relations between the main bank and the abnormal return in our sample period 1993-2003. In contrast, Kang et al. (2000) find significant positive abnormal returns for the period 1977-1993 and conclude that the main bank enhances shareholder wealth. Comparison of our findings with Kang et al. (2000) indicates that the influence of the main bank on mergers of Japanese companies appears to have changed due to the liberalization of the financial markets and dissolution of cross-shareholdings. In our second research we extend our sample period to 1982-2003 in order to examine whether the role of the main bank has indeed changed in recent years. In particular, we investigate whether the main bank enhances shareholder wealth or acts in its own interest as creditor. In this analysis we focus on the adaptive motive, mergers in which the target company is in financial distress. According to the bank power hypothesis, we can assume that a main bank of a target company with high outstanding debts and facing financial distress will be inclined to arrange a solution in this situation to protect its own interests. In relation to mergers, we assume that it will arrange a merger with a strong company to protect its own interests. As such, the main bank will not be concerned with increasing shareholder wealth. We find support for this hypothesis: when the main bank is the same for the merging companies, the merger does not create shareholder wealth. The merger is motivated by the main bank in order to protect its own interests as creditor. Generally, same main bank mergers are not between two weak companies, but at least one of the merging companies is financially strong. We find different shareholder wealth effects in the bubble period (1982-1989), the early 1990s, and the post-financial regulation regime (1997-2003). This finding indicates that the influence of external shocks is also very important in Japan.

206

218 Appendix

Appendix I - Event study Methodology

Cumulative Abnormal Return - The abnormal returns for a company can be calculated using the standard event-study methodology, following Dodd and Warner (1983). First parameters are estimated in the base period; these parameters are subsequently used in the test period to calculate residuals/abnormal returns. In this explanation we will use a base period beginning 210 days before and ending 31 days before the announcement date. The daily return data (change in share price) during this 180-day base period are used to estimate the parameters of the standard market model: (Base period = AD – 210 to AD – 31)

Rit ED ii mt eR it

where Rit is the daily return at time t of either the bidder or target company I and Rmt is the daily return at time t of the market index, the Tokyo Stock Exchange’s Stock Price Index. Alfa and beta are the parameters that will be estimated and e is the error term. If we would like to know the residuals of each bidder and target company for a 3-day test period from one day before to one day after the announcement date. The residuals for each firm, uit, are calculated by using the estimated alfa and beta as follows: (Test period = AD - 1 to AD + 1)

Ö Ru itit Ö ED ii Rmt

Average returns, Art, for the N firms on a common date t relative to the centring dates, AD and/or ED, were calculated as

and cumulative average residuals, CART, up to date T were calculated as

T CART ¦ ARt t 1

207

219 Chapter 9 Merger activity

Appendix II

Table 9-16 CAR [-5, +90] for Japanese bidders categorized by characteristics of bidder and transaction Mann- Number Mean CAR Median CAR T-Test Panel A : Main Bank and keiretsu linkages Whitney of obs (p-value) (p-value) (p-value) (p-value) All Cases 86 2.23% 2.14% (0.282) (0.161) Panel A : Main Bank and keiretsu linkages Bidder is affiliated with main bank 47 2.73% 2.85% 0.265 (0.300) (0.243) (0.792) Bidder is unaffiliated with main bank 39 1.63% 2.06% -0.004 (0.626) (0.522) (0.997) Bidder’s main bank loan ratio is above 42 3.39% 2.47% 0.547 sample median (0.228) (0.280) (0.586) Bidder’s main bank loan ratio is below 44 1.13% 2.13% 0.130 sample median (0.715) (0.451) (0.897) Bidder belongs to horizontal keiretsu 46 3.92% 3.39% 0.873 (0.149) (0.302) (0.385) Bidder independent of horizontal keiretsu 40 0.30% 1.34% 0.619 (0.926) (0.430) (0.536) PanelB:Otherfactors Bidder’s bank loan ratio is above sample 39 0.99% 1.64% -0.546 median (0.743) (0.522) (0.587) Bidder’s bank loan ratio is below sample 47 3.26% 4.40% -0.620 median (0.258) (0.243) (0.535) Bidder’s leverage ratio is above sample 44 3.33% 2.14% 0.541 median (0.195) (0.291) (0.590) Bidder’s leverage ratio is below sample 42 1.09% 1.74% 0.281 median (0.745) (0.441) (0.779) Bidder has shareholding in target 34 1.31% 4.49% -0.361 (0.717) (0.392) (0.719) Bidder does not have shareholding in 52 2.83% 1.49% 0.097 target (0.263) (0.332) (0.923) Bidder's shareholding in target is above 17 2.46% 5.19% 0.054 median (*) (0.602) (1.000) (0.957) Bidder's shareholding in target is below 69 2.18% 2.08% -0.222 median (*) (0.351) (0.148) (0.824) Bidder and target have common 34 2.56% 4.49% 0.126 shareholder (0.440) (0.058) (0.900) Bidder and target do not have common 52 2.02% 0.25% 0.711 shareholder (0.454) (0.890) (0.477) Common shareholder has shareholdings 20 4.26% 7.05% 0.537 of 5% or more in bidder and target (0.418) (0.012) (0.592) Common shareholder does not have 66 1.62% 0.25% 1.441 shareholdings of 5% or more in bidder (0.466) (0.902) (0.150)

208

220 Appendix

Table 9-16 continued Mann- Mean CAR Median CAR T-Test Number Whitney of obs (p-value) (p-value) (p-value) (p-value) Relative size (total assets target divided 41 1.59% 2.85% -0.197 by total assets bidder) is above sample (0.644) (0.211) (0.844) Relative size (total assets target divided 42 2.43% 1.86% -0.073 by total assets bidder) is below sample (0.349) (0.644) (0.942) Acquisition motivated by rescue purposes 9 -5.49% -8.65% -1.286 (0.260) (1.000) (0.202) Acquisition not motivated by rescue 77 3.14% 2.96% -1.615 purposes (0.163) (0.110) (0.106) Merger in period 1993-1997 32 -2.48% -3.02% -1.782 (0.458) (0.860) (0.078) Merger in period 1998-2003 54 5.03% 3.83% -1.845 (0.057) (0.040) (0.065) Large bidders (market cap is above 50 3.67% 4.11% 0.820 sample median) (0.143) (0.203) (0.415) Small bidders (market cap is below 36 0.24% 1.34% 0.512 sample median) (0.947) (0.618) (0.609) Target is listed 51 0.47% 2.08% -1.036 (0.872) (0.262) (0.303) Target is unlisted 35 4.81% 2.20% -0.782 (0.099) (0.500) (0.434) Bidder and target are in same industry 61 0.77% 2.06% -1.110 (0.762) (0.200) (0.270) Bidder and target in different industries 36 5.81% 4.40% -0.884 (0.105) (0.690) (0.376) Notes The sample consists of 136 Japanese bidders listed on the Tokyo Stock Exchange for which the announcement date of the merger is between 1 January 1993 and 31 December 2003. For CAR (AD-5, AD+90) we excluded cases where the effective date of the merger took place within 95 days after the announcement date, in order to avoid the influence of confounding events. We further eliminated the 20 mergers with the largest fluctuations in abnormal returns and obtained a final sample of 86 merger cases. Excess trading volume is determined by taking the average of the ratio of daily trading volume in each stock and daily trading volume in the TOPIX index over the period from -220 to -20 days before the announcement. The actual trading volume over the test period is divided by this 'normal' ratio. P-values indicate whether average excess trading volume ratio differs significantly from 1. Leverage ratio is computed as the ratio of the book value of total debt to the sum of the book value of debt and the market value of equity. Bank loan ratio is computed as the ratio of the book value of debt from financial institutions to the sum of the book value of debt and the market value of equity. Main bank loan ratio is computed as the ratio of the book value of debt from the main bank to the sum of the book value of debt and the market value of equity. P-values for t-tests, whether values differ significantly from 1, are reported in parentheses below the mean, and p-values from sign-rank tests are reported in parentheses below the median.

209

221 210

222 10

Summary and conclusions

This thesis examines M&A activity in Japan before and after the Second World War (“WWII”). The research aims to provide an overview of the developments related to M&A in Japan during these two periods. A full understanding provides us tools to interpret and understand the developments currently taking place in Japanese M&A activity. We provide information on mergers and acquisitions by listed companies in the pre-war period and a detailed examination of merger and hostile takeover activity in the post-war period.

10-1 Summary Chapter 1 introduces the research question and the structure of the thesis; it shows the thesis is divided into 4 parts and consists of 10 chapters. Chapter 2 provides a description of M&A activity in Japan over time. Chapters 3 through 5, Part II, give background information for interpretation of M&A activity. In these chapters we look into informal institutions, formal institutions, and political and economic development in the pre-war and post-war periods. We also discuss companies’ financing decision, ownership structure, corporate governance structure, and group structures. Chapters 6 and 7 of Part III primarily look into hostile takeovers in Japan. Chapter 6 introduces our institutional model for M&A that we use to analyze hostile takeovers in chapter 7. In chapter 8 we explain merger motives and drivers for merger waves, and we also discuss other types of M&A activity. Chapter 9 discusses merger activity in Japan. We look at the merger motives and drivers for merger waves in Japan, and discuss results from previous studies on mergers and the monitoring role of the main bank. Next, we execute two event studies and examine whether shareholder wealth is created around the announcement date of a merger In Chapter 2 we investigate merger and hostile takeover activity in Japan during the pre-war and post-war periods. For the pre-war period we find that M&A activity was low, not exceeding an annual total of 41 merger cases. We did not find evidence for merger waves in the pre-war period and zaibatsu companies were not more active in M&A than

211

223 Chapter 10 Summary and conclusions

non-zaibatsu companies. Merger activity was concentrated in the 1920s. In the electricity sector these were aimed at attaining economies of scale (synergy); other mergers predominantly appear to have a rescue motive. In the post-war period we find two merger waves, the first in the period 1963-72 and the second from 1991 to present. The first merger wave coincides with removal of controls on capital transactions in 1964 and appears to be partly influenced by administrative guidance of the Japanese bureaucracy. The main aim of the mergers is to prevent excessive competition and increase the scale of domestic companies in order to become competitive with foreign companies. The largest mergers are concentrated in the Steel, Transportation and Automotive sectors. The second merger wave starts after the collapse of the bubble economy and is related to industrial restructuring as the economic recession resulted in bad performance and difficulty to repay loans. In the early 1990s the largest mergers occur in the Paper, Chemical and Banking industries; mergers aimed at rescuing failing companies were also important. After the regulatory changes in the latter half of the 1990s the number and size of M&A activity increases considerably, although the number of mergers remains stable in the period 1999-2006. In this period companies initiate mergers for strategic purposes. We find a lot of hostile takeover activity in the post-war period, starting immediately after WWII. Only in 1999 the first hostile tender offer occurs, and in the period 1999 to July 2007 thirteen hostile tender offer attempts were launched. Two hostile tender offers were successful, but their “hostility” can be questioned, as the bidder company and target company had a long-standing relationship. Chapter 3 discusses informal institutions, political development, and formal institutions in Japan. Informal institutions originate from Shinto, Buddhism, and Confucianism. Important virtues related to mergers and hostile takeovers are harmony, tolerance, modesty, importance of the group, and trust. Other important social informal institutions in Japan are the ie and dǀzoku systems, which represent the relationship between individuals and companies in a group. In the pre-war period Japan was determined to catch up with the West and to amend the unequal treaties. This had an important influence on the formal institutions in Japan; a modern banking system, stock market and bond market were set up and functioned effectively up to 1937 when Japan turned into a military state. A constitution was promulgated that changed the political system into a centralized bureaucratic state. After the war, Supreme Commander for the Allied Powers’ policies were aimed at demilitarizing and democratizing Japan and setting up a regulatory environment for a safe and stable financial system. A new constitution was enacted and changed Japan into a democratic country. The stock market and the bond market are characterized by deregulation in the 1980s and 1990s. The change in laws related to ownership was most important for the ownership structure, governance structure, and M&A activity after the war. Prior to WWII the general shareholders’ meeting had unlimited authority in all financial and managerial matters. After WWII, the general shareholders’ meeting lost this power and the minority shareholders gained in influence. Authority on managerial matters was transferred to the board of directors. Other important changes after WWII were the ban on establishment of pure holding companies and prohibition of shareholdings between companies of the same pre-war zaibatsu group. This latter law was abandoned after the peace treaty was signed in 1952, and the pure holding company is permitted again as of 1997. Various laws directly

212

224 Section 10-1 Summary

related to M&A were implemented in the late 1990s, resulting in simplified merger procedures, merger currency flexibility, and easier notification procedures for tender offers. As of 2001 it is possible for companies to buyback their own shares and cancel or hold them as treasury stock. Chapter 4 examines the influence of informal and formal institutions on the financing decision, ownership structure, and governance structure in the pre-war and post-war periods. In this chapter we investigate aggregate data, and in chapter 5 we look into a case study of Mitsubishi groups. To complement the discussion on political development and formal institutions, a brief overview of the economic development in Japan as of 1868, the start of the Meiji period (1868-1912), is given. The government took the initiative in industrial development and established companies in various industrial sectors. The government feared to become colonized by Western countries and realized that individual businessmen did not have sufficient wealth to make the necessary investments. In the 1880s the government sold its companies to businessmen with political relations (seishǀ) and ability to build strong and internationally competitive companies. These businessmen became the leaders of the pre-war zaibatsu groups, conglomerates of various companies, organized in pyramid form, owned by rich families and/or a family-owned holding company. In the post-war period government involvement changed into administrative guidance of companies. New laws prevented establishment of pure holding companies, and new corporate structures developed in the post-war period: the vertical keiretsu and horizontal keiretsu. Corporate governance in the post-war period is characterized by the main bank system. We find that in both the pre-war and the post-war period the trade association played an important role. Companies in the same industrial sector worked closely together, discussed strategies, shared technical information, protected members’ privileges, and established trade rules. The most important method of financing in the pre-war period was internal finance, but in the 1930s external finance increased in importance. We show that only a gradual decline in equity finance started as of 1942. In the post-war period the stock and bond markets were not efficient and expensive for capital procurement. This led companies to rely to a high degree on bank loans for financing. We look into the developments of share- ownership in the 1950s and show that individuals were unable to purchase new shares; this finding is in contrast to the commonly used argument that individual shareholders sold their shares during this period. Deregulation of financial markets in the 1980s has resulted in more financing opportunities for companies and a weakening position for the main bank. The recession in the 1990s and the banking crisis of 1997 also resulted in unwinding of shareholdings in the horizontal keiretsu. Chapter 5 analyzes the corporate groups that were discussed in Chapter 4 for the Mitsubishi group. For the pre-war period we look into the Mitsubishi zaibatsu. Before the war the Mitsubishi holding company and Iwasaki Family were the most important shareholders, owning in most cases more than 90% of the shares in the main group companies. During the 1930s the zaibatsu families started to change the ownership structure due to anti-zaibatsu pressures, and the first group shareholdings emerged. Mitsubishi Bank and Mitsubishi Trust Bank were relatively unimportant shareholders. During the war period 1937-45, when loan financing increased, banks did not develop into important shareholders of zaibatsu companies as well.

213

225 Chapter 10 Summary and conclusions

In the post-war period the following corporate groups developed: Mitsubishi horizontal keiretsu, the main bank system with the Mitsubishi Bank at the apex, and the Mitsubishi vertical keiretsu. Financial institutions play an important role in loan provision and share ownership of group companies in the first two groups. The number of cross-shareholdings between companies in horizontal keiretsu is very low and concentrated in six companies. We term the six companies as the ‘Mitsubishi keiretsu diamond’ companies, and these have close relations with financial institutions through cross-shareholdings. The keiretsu diamond companies and financial institutions have extensive ownership in other (important) group companies. The diamond companies and the four financial institutions are all direct subsidiaries of the pre-war Mitsubishi holding company and Iwasaki family. This indicates that the pre-war apex companies remain important as apex companies in the post-war period. Companies in the horizontal keiretsu are in most cases also the core company in a vertical keiretsu, but some companies in the horizontal keiretsu are part of the vertical keiretsu of another horizontal keiretsu company. We argue that it is possible that findings in previous research on the horizontal keiretsu are biased due to the influence of the relations between a parent company and a subsidiary in a vertical keiretsu. Chapter 6 presents the institutional model we developed to analyze M&A activity in Japan. The model is inspired on the Williamson 4-level model (Williamson 1998), Williamson 3-level model (Williamson 1996), and insights from North (1990, 2005) and Greif (2006). The model emphasizes the importance of motivation of actors and the historical and context-specific background of institutional elements. The four levels of the model are as follows: informal institutions at level one, formal institutions at level two, governance structure and M&A activity at level three, and individual actors at level four. The informal institutions constrain all other levels, and formal institutions constrain governance structure, M&A activity, and individual actors. M&A activity, governance structure, and individual actors are linked with bi-directional solid arrows; they influence, and are influenced by, each other. M&A activity results from (strategic) considerations of individual actors, such as, for example, a company’s management, its shareholders, or a hostile raider. The governance structure of a company will have an important impact on these considerations and wether they will succeed. At the same time, M&A activity affects the governance structure, and can have an impact on the behaviour of individual actors in relation to stock ownership and M&A activity. Level three is the outcome of the dynamic interactions between economic development, (foreign) competition, and changes in informal and formal institutions that govern motivation and behaviour of individual actors. The neoclassical black boxes of governance structure and individual actors are opened; they are influenced by each other, by M&A activity, and by informal and formal institutions. Governance structure consists of two layers: the company and corporate groups. The box of individual actors is made up of three layers: (i) shared cognition, (ii) individual mental maps and shared mental maps, and (iii) motivation. In our model, institutional change is a historical process and needs to incorporate historical development of institutions and motivation of actors. Chapter 7 analyzes hostile takeovers in Japan. We first discuss arguments from previous research and then provide our own interpretation based on our institutional model for M&A. In previous research the horizontal keiretsu is argued to be formed as a result of hostile takeovers in the 1950s; this corporate group has subsequently prevented hostile

214

226 Section 10-1 Summary

takeovers in the entire post-war period. Another argument is that hostile takeovers do not fit in with the Japanese culture. According to transaction cost economics, hybrid structures without contractual safeguards can be sustained in Japan because of a high level of trust, importance of reputation, and financial hostages. With our institutional model, we examine post-war hostile takeover activity in Japan. First, we look into the concept of trust and the historical development of corporate groups. The Japanese society is characterized by high particularistic trust and low general trust. The horizontal keiretsu is an example of the ie system and has its origins in the pre-war zaibatsu, as a result of the new formal institutions implemented after WWII. The vertical keiretsu developed after WWII and is characterized by a relationship in which the parent company and the subsidiary are mutually dependent. Trade associations originate from 1092AD and have been very important in disseminating new technology and information to companies in the same business sector and in influencing government policy. In section 10-2 we will explain their influence on hostile takeovers according to our institutional model for M&A. Chapter 8 provides an overview of the various types of M&A activity and distinguishes activities into those related to (i) diversification or expansion, and (ii) restructuring. The main focus of the thesis is diversification or expansion activities structured as mergers and acquisitions (in particular hostile takeovers). Regarding hostile takeovers, in the form of tender offers, the chapter gives some examples of available defensive measures such as the poison pill, white knight, white squire, and green mail. The most important merger motives for companies are synergy, adaptive, managerial empire building, and hubris. Merger motives refer to individual M&A transactions, drivers for merger waves to more general aspects that induce a lot of companies to engage in M&A activity in a particular period. Two important drivers for merger waves are external shocks and overvaluation of stocks. Both the drivers for merger waves and merger motives were different for the five merger waves that occurred in the United States. The first wave (1900) was aimed at increasing market power, the second wave focused on restructuring industrial sectors. The third wave in the 1960s was aimed at diversification and creation of large conglomerates, whereas the fourth wave was dominated by mergers between large companies and hostile takeovers. The fifth wave had strategic motives. Chapter 9 discusses our findings on M&A activity from chapter 2, results from previous research on M&A in Japan, and our own empirical research. In the first section of this chapter, we relate M&A activity in Japan to merger motives and drivers for merger waves. In the pre-war period we did not find merger waves. Pre-war mergers in the electricity sector appear to be motivated by synergy; and in other sectors the adaptive motive seems to be prevalent; a lot of companies fell in financial distress during the 1920s, and mergers appear to have been aimed at rescuing failing companies. Similar to the post- war period, mergers were arranged by the target company or a third party, such as a bank. In the 1960s, during the first merger wave, mergers were aimed at preventing excessive competition and achieving economies of scale in order to be able to compete with foreign companies (synergy motive). The first period of the second wave (1991-1996) is dominated by mergers between companies in financial difficulty due to the economic business cycle (adaptive motive). In the second period, as of 1997, we see that companies are engaging in mergers for strategic purposes. Next, we discuss previous research on (i) main bank monitoring, (ii) abnormal returns at the announcement of mergers in Japan and the U.S., and (iii) profitability drivers of

215

227 Chapter 10 Summary and conclusions

mergers. Subsequently we perform two empirical event studies that examine the stock market reaction around the announcement date of a merger. In the first empirical study we look into the abnormal return of bidder companies between 1993 and 2003 in 136 domestic mergers between non-financial companies in Japan. The sample includes listed as well as unlisted target companies. We do not find a positive association of affiliation with a main bank or keiretsu membership with the announcement returns. On the other hand, the presence of a common shareholder holding shares in both the bidder and the target company, and the period in which the merger took place (in/after 1998), have a positive influence. In the second empirical study we examine the abnormal return of listed bidder companies and listed target companies in 91 mergers between non-financial companies in the period 1981 to 2003. In this research we look in more detail into the relation of the merging companies’ abnormal return, their affiliation with the same main bank, whether they are in financial distress, and their profitability. We find that the companies with the same main bank had high profitability in the 1980s, but that the main bank arranged bail- outs for failing target companies in the 1990s.

10-2 Conclusions In this section we come back to our central question and sub-questions. We first discuss the determinants of M&A activity in Japan, in particular merger motives and the role of the main bank, and then look into hostile takeovers.

Central question: What are the determinants of M&A activity in Japan over time?

Sub-question I Is hostility in M&A activity increasing, and which actors initiate hostile takeover activities?

Sub-question II What motivates Japanese companies to engage in mergers, which actors play a role, what are the profitability drivers, and is shareholder wealth created?

Determinants of M&A activity in Japan This thesis shows that for an analysis of M&A activity in Japan, it is necessary to include the influence of various elements, starting with political and economic development and informal and formal institutions. These institutions have an important impact on actors and governance structures of companies. Informal and formal institutions influence companies’ ownership structure, governance structure, and M&A activity. The impact of these institutions is through the behaviour of actors. Formal institutions, constrained by informal institutions, affect merger activity through the actors as well. The informal and formal institutions need to be considered in relation to the motivation of actors to follow particular rules. This argument also applies with reference to the financing decision and available methods to procure funds. The ownership and governance structure of a company have a direct influence on M&A activity. The management’s financing decision and a company’s ownership structure and

216

228 Section 10-2 Conclusions

corporate governance structure determine the relative influence of shareholders and creditors on M&A decisions. Merger activity is also influenced by various elements such as (international) competitiveness, government regulations, changing domestic and international political-economic conditions, and industrial policies. In order to understand behaviour of actors and governance structures, it is indispensable to reflect on the historical and context-specific background. The resulting behaviour of actors and/or governance structure following a change in, for example, formal institutions, can only be understood when the historical background is taken into account. The behaviour of actors is the result of the interactions of an actor’s (i) shared cognition, (ii) individual mental maps and shared mental maps, and (iii) motivation. The behaviour of the actors, constrained by the informal and formal institutions, results in the governance structure and M&A activity. This is exemplified by our finding that the horizontal keiretsu that developed after WWII was the consequence of the new company law, rather than caused by the threat of hostile takeovers. Or, to put it differently, the origin of the hostile takeovers threat lies in the new company law that significantly changed the authority distribution of shareholders. Behaviour of managers can also be seen in the merger motives of M&A activity. There are various motives for mergers, such as synergy, adaptive, managerial empire building, and hubris. The merger motive refers to an individual M&A transaction, whereas merger wave drivers represent more general aspects that induce a lot of companies to engage in M&A activity, such as external shocks and overvaluation of stocks. In our analysis of M&A activity in the pre-war and post-war periods, we found that in both periods the main motives for mergers were the synergy and adaptive motives. As a result of the informal institutions, Japanese managers do not appear to engage in mergers with a managerial empire building or hubris motive. We also find that corporate groups have an important influence on M&A activity. As we will show below in our discussion on hostile takeovers, the vertical keiretsu and trade association appear to have an important influence in Japan. Regarding mergers we look into the main bank system, as the main bank is argued to have an important monitoring role over Japanese companies. We relate this to the adaptive motive of mergers and examine what kind of rescue mergers occurs in Japan - and whether the relationship of a main bank influences merger activity. We assume that the main bank of a company in financial distress will attempt to arrange a merger of a weak company with a stronger company. The merger will protect the main bank’s interests as a creditor and not necessarily be intended to increase shareholder wealth. In mergers between companies with the same main bank, we find that shareholder wealth is not created, and that the main bank acts to protect its own interests as creditor.

Hostile takeovers Previous research argues that the low level of hostile takeover activity in Japan is primarily caused by two institutional elements in the Japanese society. The first is the horizontal keiretsu, which is argued to have prevented hostile takeovers through stable and cross- shareholdings between companies. The second is the specific Japanese culture that is said to be characterized by the pursuit of harmony. This thesis provides evidence that the trade association and the vertical keiretsu might play an important role in hostile takeovers. We conclude that assumptions made on the influence of the other two Japanese characteristics might need to be reconsidered. Our main contribution to the academic discussion on M&A in Japan is that we show that institutional elements need to be looked at from a broad and

217

229 Chapter 10 Summary and conclusions

historical perspective. An institutional element can only be understood if perceived in relation to other institutional elements and taking into account its dynamic development over time, rather than in isolation and at one particular point in time. We show that the horizontal keiretsu, as the most important prevention measure against hostile takeovers needs further exploration. Based on our analysis, we conclude that in the post-war period all hostile takeovers need to be interpreted as greenmail attempts. New laws implemented after WWII created the possibility to engage in greenmail. As companies could not repurchase their own shares before 2001, companies that were part of a horizontal keiretsu were targeted. Rather than preventing hostile takeovers, the horizontal keiretsu appear to have caused greenmail in the post-war period. The argument that the Japanese, as a people, put high emphasis on harmony and trust, implies that they will not undertake hostile actions. We show that there are two types of trust: general trust and particularistic trust. General trust is trust between individuals that do not have a particular relationship, and particularistic trust refers to trust in groups of mutually dependent individuals. The Japanese society is characterized by low general trust and high particularistic trust. The low general trust can be seen in the greenmail practices towards companies, the high particularistic trust in, for example, trade associations. Trade associations are formed by companies in the same industrial sector: information is exchanged and relationships are built. We show that because of the strong relations in these groups, companies build a high level of particularistic trust, preventing these companies from engaging in hostile takeovers vis-à-vis each other. The vertical keiretsu is another important impediment to hostile takeovers, functioning in three ways. The first characteristic is that a large Japanese company is not the same as a large U.S. company. Whereas U.S. companies tend to produce most products in-house, Japanese companies deploy a wide subcontracting system that makes it a difficult target for a hostile takeover. The second characteristic of the vertical keiretsu is that supply from the subsidiaries is indispensable in the production process of the parent company. The majority of its important subsidiaries is already fully owned, but if a hostile takeover is attempted on an important, partially owned subsidiary, the parent company will intervene to secure its own production process. The third aspect is the question whether it would be beneficial for a hostile bidder to try and takeover an unwilling subsidiary company. Even if the hostile bidder would succeed, it is very likely that it will lose the subsidiary’s suppliers and customers in the process. These factors may have limited hostile takeover attempts to a high degree.

Regarding our sub-question whether hostility in Japanese M&A is increasing, we do not believe this is the case. Since WWII, Japanese companies have had to cope with a lot of hostilities from outside the company, and only the approach of the raider appears to be changing. Although there seems to be an increase in hostility in Japan, this is primarily caused by the fact that hostilities are engaged in by respectable investment companies and deals are publicly announced. This is different from the silent greenmail by shite groups, but the motivation appears to be the same. This, combined with the fact the attempts remain at a very low level, leads us to question the generally accepted explanations and the influence of the unwinding of cross-shareholdings and changes in Japanese cultural characteristics. With our model we therefore propose the vertical keiretsu and trade association as explanations why Japanese blue-chip companies do not initiate hostile tender offers.

218

230 Section 10-2 Conclusions

We think it is only possible to argue that hostility has increased, if and when Japanese blue-chip companies start launching hostile tender offers on companies on a regular basis. To our opinion, a different hostile takeover market in Japan can only result from the weakening of relations within the vertical keiretsu and trade association. As this will not occur in the near future, we expect that only (foreign) investment funds will be active players. Resulting from the above-mentioned institutional elements, however, they will not be able to gain operational control of a company or use the target for asset-stripping. As hostile tender offers will be prevented by a company’s buyback of shares, poison pills or mergers and/or acquisitions by friendly (related) companies, the funds will be able to achieve significant financial gains.

Institutional model for M&A revisited Our model indicates that a lot of different aspects have influenced, and still influence, M&A activity in Japan. The question we need to ask; will the model be useful in the future? We think it will, primarily as it includes the important element of the historical and context-specific background. So, by definition it will be applicable for future studies. On the other hand, as previous studies never included the vertical keiretsu and trade association in the discussion of hostile takeovers in Japan, there might be elements that are not included in our model that have an important influence on hostile takeovers. For example, we did not look into the decision-making process related to new M&A regulations. The sudden extension of implementation of the triangular mergers from May 2006 to May 2007, indicates that important processes occur between managers and law- makers. Also, the role of foreign investors might need to be looked into in more detail to understand their influence in the future. We think that the new elements of our model are indispensable in any discussion of hostile takeovers in Japan, but, depending on new insights in the future, new elements may have to be added.

219

231 220

232 Dutch summary

In dit proefschrift worden fusies en overnames (M&A) in Japan onderzocht, waarbij de nadruk ligt op (i) de rol van de ‘main bank’ in fusies en (ii) greenmail en vijandige overnames in de naoorlogse periode. In Japan is het aantal fusies tussen grote bedrijven laag en zijn vijandige overnames zeldzaam. Sinds de tweede helft van de jaren ‘90 stijgen het aantal en de totale waarde waarde van M&A transacties. In overeenstemming met de trend in de totale activiteiten op het gebied van M&A komen vijandige tender offers ook vaker voor: in de naoorlogse periode tot 1999 waren er geen vijandige tender offer pogingen, maar in de periode van 1999 tot juli 2007 waren er dertien. Het knappen van de zeepbeleconomie en de daaropvolgende recessie in de jaren ‘90 zetten Japanse ondernemingen aan na te denken over de efficiëntie van het traditionele Japanse systeem van corporate governance. Ook reduceerde de deregulering van de financiële markten de invloed van de main bank en resulteerden de crisis in de financiële sector en de nieuwe financiële verslaggevingsregels in de tweede helft van de jaren ‘90 in het tot ontwikkeling komen van stabiel en wederzijds aandelenbezit van ondernemingen. Specifieke Japanse culturele karakteristieken die zijn gebruikt ter verklaring van het lage aantal M&A transacties, zijn ook veranderd en dit leidde tot herstructurering van bedrijven. Dit proefschrift onderzoekt hoe en waarom M&A activiteit in Japan is veranderd. Wij bouwen een uniek gegevensbestand op over M&A activiteit in het vooroorlogse en naoorlogse Japan. De analyse van de vooroorlogse periode betreft 537 ondernemingen die genoteerd zijn aan de effectenbeurs van Tokyo (TSE). Wij tonen dat M&A activiteit erg laag is in de periode 1906-37 en dat er geen bewijs is voor veel vijandige overnames. Ook geven wij aan dat bedrijven gerelateerd aan een zaibatsu groep niet actiever in M&A waren dan bedrijven zonder een dergelijke relatie. In de naoorlogse periode vinden wij twee fusiegolven: de eerste golf in de periode 1963-1972 en een tweede golf die begint in 1990 en tot op heden voortduurt. De eerste golf werd veroorzaakt door de handelsliberalisering en de opheffing van restricties op kapitaaltransacties in 1964. Deze fusies waren gericht op het voorkomen van hevige concurrentie tussen Japanse bedrijven en het versterken van hun mondiale concurrentiepositie. De tweede golf begint na het knappen van de zeepbeleconomie en is gerelateerd aan bedrijfsherstructurering. Na de wijzigingen in de wetgeving in de jaren ‘90 nemen het aantal en de grootte van M&A transacties toe, maar het aantal fusies blijft relatief laag en stabiel gedurende de periode 1999-2006. In tegenstelling tot de Amerikaanse fusiegolven, zijn de golven in Japan anticyclisch, met betrekking tot zowel de algemene economie als de aandelenkoersen. Om vijandige overnames te onderzoeken, bouwen wij een institutioneel model voor M&A activiteit waarin wij institutionele elementen zoals cultuur, wetten, aandelenbezit en groepsstructuren van ondernemingen integreren. Eerder onderzoek stelt dat het lage aantal vijandige overnames in Japan is veroorzaakt door twee institutionele elementen in Japan. De eerste is de horizontale keiretsu: vijandige overnames zouden worden voorkomen door stabiel en wederzijds aandelenbezit tussen ondernemingen. De tweede is de specifieke Japanse cultuur, met als een van de belangrijkste kenmerken het nastreven van harmonie en vertrouwen. Wij tonen met ons model aan dat de bewering dat de horizontale keiretsu de belangrijkste factor is geweest ter voorkoming van vijandige overnames, twijfelachtig is. Op basis van onze analyse concluderen wij dat alle vijandige overnames in de naoorlogse

221

233 Dutch summary

period als greenmail moeten worden beschouwd. Nieuwe wetten in de naoorlogse periode creërden de mogelijkheid tot greenmail. Voor 2001 was het voor Japanse ondernemingen niet mogelijk om hun eigen aandelen in te kopen en dit leidde tot greenmail gericht op bedrijven die deel uitmaakten van een horizontale keiretsu. Wij zijn van mening dat horizontale keiretsu niet zozeer vijandige overnames hebben voorkomen, als wel greenmail hebben veroorzaakt. Het argument dat Japanners erg veel waarde hechten aan harmonie en vertrouwen, impliceert dat Japanners geen vijandige acties zullen initiëren. Wij tonen aan dat ‘vertrouwen’ moet worden opgesplitst in algemeen vertrouwen en specifiek vertrouwen. Algemeen vertrouwen is het vertrouwen tussen individuen die geen specifieke relatie met elkaar hebben; specifiek vertrouwen duidt op het vertrouwen binnen groepen individuen die van elkaar afhankelijk zijn. De Japanse samenleving wordt gekarakteriseerd door een laag algemeen en een hoog specifiek vertrouwen. Het lage algemeen vertrouwen manifesteert zich in de greenmail van ondernemingen, het hoge specifiek vertrouwen in bijvoorbeeld de bedrijfsorganisatie. Wij concluderen uit ons model dat bedrijfsorganisaties en verticale keiretsu wellicht een belangrijke rol spelen met betrekking tot vijandige overnames. Een bedrijfsorganisatie wordt gevormd door ondernemingen uit dezelfde industriële sector; er wordt informatie uitgewisseld en er worden relaties opgebouwd. Wij tonen aan dat ondernemingen door de sterke relaties binnen deze groepen een hoog niveau van specifiek vertrouwen opbouwen, waarmee voorkomen wordt dat ondernemingen vijandige overnames tegen elkaar initiëren. De verticale keiretsu is een andere invloedrijke factor bij vijandige overnames. Deze functioneert op drie manieren. Op de eerste plaats is een grote Japanse onderneming niet hetzelfde als een grote Amerikaanse onderneming. In een Amerikaanse onderneming worden de meeste producten binnen de onderneming zelf gemaakt, maar een Japanse onderneming besteedt haar productie op grote schaal uit. Dit maakt een Japanse onderneming een moeilijk object voor een vijandige overname. Het tweede kenmerk van de verticale keiretsu is dat de producten van de onderaannemers onmisbaar zijn in het productieproces van de moedermaatschappij. De meerderheid van de onderaannemers is al volledig in bezit van de moedermaatschappij. Indien er echter een poging tot vijandige overname van een gedeeltelijk in bezit zijnde onderaannemer wordt gedaan, zal de moedermaatschappij ter bescherming van haar eigen productieproces actie nemen om dit te voorkomen. Het derde aspect is de vraag of het gunstig is voor een vijandige aspirant- overnemer om te proberen een onderaannemer over te nemen als deze er onwillend tegenover staat. Zelfs als de vijandige aspirant-overnemer in zijn opzet zou slagen, is het zeer waarschijnlijk dat hij in de loop van het vijandige overnameproces de leveranciers en klanten van de overgenomen onderneming zal verliezen. Deze factoren hebben vijandige overnames in hoge mate beperkt. Wij zijn van mening dat alleen verzwakking van de relaties in de verticale keiretsu en bedrijfsorganisaties zal resulteren in veranderingen in vijandige overnames in Japan. De fusies die plaatsvinden gedurende de tweede fusiegolf creëren geen aandeelhouderswaarde. Wij voeren twee eventstudies uit en onderzoeken de invloed van Japanse kararakteristieken zoals de main bank, horizontale keiretsu en de aanwezigheid van een moedermaatschappij. Ook kijken wij naar bedrijfsgerelateerde gegevens als bijvoorbeeld de winstgevendheid en liquiditeitsproblemen. Op de dag van de aankondiging van een fusie vinden wij geen positieve correlatie tussen de koers die een aspirant overnemende partij biedt en de relatie die deze partij heeft met een main bank of

222

234 Dutch summary

horizontale keiretsu. Als de aspirant overnemende partij (bieder) en de overnamekandidaat dezelfde main bank hebben, creërt de betrokkenheid van de main bank geen aandeelhouderswaarde maar is haar eigen belang als crediteur de motivatie: fusies waarbij een main bank een rol speelt, vinden niet plaats tussen twee zwakke bedrijven, maar tussen bedrijven waarvan tenminste één bedrijf financieel sterk is. Begin jaren ‘90 vonden de meeste fusies plaats tussen een zwakke overnamekandidaat en een sterke bieder.

223

235 224

236 References

Abegglen, J. C. 1983. Why Japan Mistrusts Mergers. Mergers & Acquisitions 17 (4): 16- 19. Abegglen, J.C and G. Stalk. 1985. Kaisha, the Japanese Corporation.NewYork:Basic Books, Inc, Publishers. Akimoto, H. 1993. The Stock Market (II). In FAIR fact series: Japan’s financial markets. Japan: Foundation for Advanced Information and Research. Allen G.C. 1946. A short Economic History of Modern Japan 1867-1937. London: Unwin Brothers Limited. Andrade, G., and E. Stafford. 2004. Investigating the economic role of mergers. Journal of Corporate Finance 10: 1–36. Aoki, M and P. Sheard.1992. Corporate Monitoring and Governance in the Main Bank System. Center for Economic Policy Research Discussion Paper. Stanford University. Aoki, M. 1988. Information, Incentives, and Bargaining in the Japanese Economy. Cambridge: Cambridge University Press. Aoki, M. 1990. Toward an Economic Model of the Japanese Firm. Journal of Economic Literature 8: 1-27. Aoki, M. 1994. “Monitoring Characteristics of the Main Bank System: An analytical and developmental view”. In M. Aoki and H. Patrick, eds., The Main Bank System.New York: Oxford University Press. Aoki, M. and H. Patrick, eds. 1994. The Japanese Main Bank System: Its Relevance for Developing and Transforming Economies. Oxford: Oxford University Press. Aoki, M., H. Patrick and P. Sheard. 1994. “The Japanese Main Bank System: An introductory overview”. In M. Aoki and H. Patrick, eds., The Japanese Main Bank System: Its Relevance for Developing and Transforming Economies. Oxford: Oxford University Press. Aoyama , H., and T. Nishikawa. 1958. Business Fluctuations in Japanese Economy during the Inter-war Period. Kyoto University Economic Review 28(1): 14-39. Arikawa, Y. and H. Miyajima. 2007. Understanding the M&A boom in Japan: What drives Japanese M&A? RIETI Discussion Paper 07-E-042. Aristotle. 350BC. Nicomachean Ethics. Translated by W. D. Ross. Asanuma, B. 1989. Manufacturer-Supplier Relationships in Japan and the Concept of Relation-Specific Skill. Journal of the Japanese and International economies 3: 1-30. Asanuma, B. 1994. “Co-ordination between Production and Distribution in a Globalizing Network of Firms: Assessing Flexibility Achieved in the Japanese Automobile Industry”. In M. Aoki and R. Dore, eds. The Japanese Firm. The Sources of Competitive Strength. Oxford: Oxford University Press: 117-153. Asquith, P. 1983. Merger Bids, Uncertainty, and Stockholder Return. Journal of Financial Economics 11: 51-84. Asquith, P., R.F. Bruner and D.W. Mullins, Jr. 1983. The Gains to Bidding Firms from Merger. Journal of Financial Economics 11: 121-139. Beasley, W.G. 1963. The Modern History of Japan. New York, London: Frederick A. Praeger, Inc., Publisher.

225

237 References

Berger, P. and E. Ofek. 1995. Diversification’s Effect on Firm Value. Journal of Financial Economics 37: 39-65. Berglof, E. and E. Perotti. 1994. The governance structure of the Japanese financial keiretsu. Journal of Financial Economics 36: 259-284. Berkovitch, E. and M.P. Narayanan. 1993. Motives for Takeovers: An Empirical Investigation. Journal of Financial and Quantitative Analysis 28 (3): 347-362. Blakemore, T.L. and M. Yazawa. 1953. Japanese Commercial Code Revisions Concerning Corporations. American Journal of Comparative Law (2): 12-24. Bradley, M., A. Desai and E.H. Kim. 1988. Synergistic Gains from Corporate Acquisitions and Their Division Between the Stockholders of Target and Acquiring Firms. Journal of Financial Economics 21 (1): 3-40. Brealey, R., and S. Myers. 1996. Principles of Corporate Finance,5th ed. New York: McGraw-Hill. Bruner, R. 2002. Does M&A Pay? A Survey of evidence for the Decision-Maker. Journal of Applied Finance 12 (1): 48-68. Bruner, R.F. 2004. Applied mergers and acquisitions.Hoboken,NewJersey:JohnWiley & Sons, Inc. Caldarola, C. 1965. Socio-Economic Dualism in Japan. Monumenta Nipponica 20 (No.3/4): 359-373. Campbell J.L. and L. Lindberg. 1991. The Evolution of Governance Regimes. In J. Campbell, J.R. Hollingsworth and L. Lindberg, eds., Governance of the American Economy. Cambridge: Cambridge University Press. Campbell, J.L. 1997. “Mechanisms of Evolutionary Change in Economic Governance: Interaction, Interpretation and Bricolage”. In L. Magnusson and J. Ottoson eds., Evolutionary Economics and Path Dependence. Cheltenham, UK, Brookfield, US: Edward Elgar. Campbell, J.Y. and Y. Hamao. 1994. “Changing patterns of Corporate Financing and the Main Bank System in Japan”. In M. Aoki and H. Patrick, eds., The Main Bank System. New York: Oxford University Press. Caves, R.E. and M. Uekusa. 1976. Industrial organization in Japan. Washington, D.C.: The Brookings Institution. Chang, S.Y. 1998. Takeovers of privately held targets, methods of payment, and bidder returns. Journal of Finance 53: 773-784. Christelow, D. 1995. When Giants Converge: The Role of U.S.-Japan Direct Investment. Armonk, NY: M.E. Sharpe. Clark, K. and E. Ofek. 1994. Mergers as a Means of Restructuring Distressed Firms: An Empirical Investigation. Journal of Financial and Qualitative Analysis 29: 541-565. Coase, R.H. 1937. The Nature of the Firm. Economica IV: 385-405. Cole, R.E. 1971. Japanese Blue-Collar: The Changing Tradition,UniversityofCalifornia Press, Berkeley, CA. Comanor, W.S. 1967. Market structure, product differentiation, and industrial research. The Quarterly Journal of Economics 81(4): 639-657. Comment, R. and G.A. Jarrell. 1995. Corporate Focus and Stock Returns. Journal of Financial Economics 37: 67-87. Commons, J.R. 1932. The Problem of Correlating Law, Economics, and Ethics. Wisconsin Law Review 8: 3-26.

226

238 References

Commons, J.R. 1961 (1934 reprint). Institutional Economics: Its Place in Political Economy. Madison: The University of Wisconsin Press. Commons, J.R., S. Perlman, and K.H. Parsons. 1950. The Economics of Collective Action. New York: Macmillan Company. Conyon, M.J. and K.J. Murphy. 2002. Stock based compensation. In J.A. McCahery, P. Moertland, T. Raaijmakers and L. Renneboog, eds., Corporate Governance Regimes: Convergence and Diversity. Oxford: Oxford University Press. Cornell, J.B. 1964. Dozoku: An Example of Evolution and Transition in Japanese Village Society. Comparative Studies in Society and History 6 (4): 449-480. Crump, J. 2003. Nikkeiren and Japanese Capitalism. London, New York: RoutledgeCurzon. Cusumano, M.A. 1985. The Japanese Automobile Industry: Technology and Management at Nissan and Toyota. Cambridge, Mass.: Harvard University Press. Denzau, A.T. and D.C. North. 1994. Shared Mental Models: Ideologies and Institutions. Kyklos 47(1): 3-31. Dewey, D. 1961. Mergers and Cartels: Some Reservations about Policy. American Economic Review 51: 255-262. Dodd, P. 1980. Merger proposals, management discretion and stockholder wealth. Journal of Financial Economics 8: 105-37. Dodd, P. and J.B. Warner. 1983. On corporate governance: A study of proxy contests. Journal of Financial Economics 11, 401-438. Dore, R. 1983. Goodwill and the Spirit of Market Capitalism. British Journal of Sociology (34): 459-482. Dore. R. 1986. Flexible Rigidities: Industrial Policy and Structural Adjustment in the Japanese Economy, 1970-1980. Stanford: Stanford University Press Dyer, J.H. 1996. Does Governance Matter? Keiretsu Alliances and Asset Specificity As Sources of Japanese Competitive Advantage. Organization Science 7 (6): 649-666. Eggertson, T. 1990. Economic Behavior and Institutions. Cambridge: Cambridge University Press. Eggertsson, T. 1996. “A Note on the Economics of Institutions”. In L.J. Alston, D.C. North and T. Eggertsson, eds., Empirical Studies in Institutional Change.NewYork: Cambridge University Press. Eis, C. 1969. The 1919-1930 Merger Movement in American Industry. Journal of Law and Economics 12 (2): 267-96. Faccio, M., J.J. McConnell and D. Stolin. 2004. When Do Bidders Gain? The Difference in Returns to Acquirers of Listed and Unlisted Targets. http://ssrn.com/abstract=467488. Flath, D. 2000. The Japanese economy. Oxford: Oxford University Press. Friedman, D.B. 1988. The misunderstood miracle : industrial development and political change in Japan. Ithaca: Cornell University Press. A Cornell study in political economy. Fruin, W.M. 1992. The Japanese Enterprise System. New York: Oxford University Press Fukasaku, Y. 1992. Technology and industrial development in pre-war Japan: Mitsubishi Nagasaki Shipyard 1884-1934. London: Routledge. Gaughan, P.A. 1999. Mergers, Acquisitions, and Corporate Restructurings. Hoboken, New Jersey: John Wiley & Sons, Inc. Gerlach, M. 1987. Business Alliances and the Strategy of the Japanese Firm. California Management Review 30 (1): 126-42. Goergen, M. and L. Renneboog. 2003. Shareholder Wealth Effects of European Domestic

227

239 References

and Cross-Border Takeover Bids. ECGI Working Paper Series 08/2003. Greif, A. 2006. Institutions and the path to the modern economy : lessons from medieval trade. Cambridge: Cambridge University Press. Groenewegen, J. and J. Vromen. 1997.” Theory of the Firm Revisited: New and Neo- institutional Perspectives”. In L. Magnusson and J. Ottoson, eds., Evolutionary Economics and Path Dependency. Cheltenham(UK)/Brookfield (US): Edward Elgar. pp.33-56. Hadley, E.M. 1970. Antitrust in Japan. Princeton, New Jersey: Princeton University Press. Harford, J. 2005. What Drives Merger Waves? Journal of Financial Economics 77: 529- 560. Harris, R.S., J.F. Stewart and W.T. Carleton. 1982. “Financial Characteristics of Acquired Firms”.InM.KeenanandL.J.White,eds., Mergers and Acquisitions: Current Problems in Perspective. Lexington, MA: Lexington Books. Hart, J.A. 1992. Rival Capitalists: International Competitiveness in the United States, Japan, and Western Europe. Cornell University Press. Hasegawa, H. 1996. The Steel Industry in Japan: A Comparison with Britain. London, New York: Routledge. Hattori, H. 2004. Jissen M&A Management [The Practice of M&A Management] (in Japanese). Toyo keizai shinposha. Hayek, F.A. 1967. Studies in Philosophy, Politics and Economics. Chicago: University of Chicago Press. Hayward, M.L.A. and D.C. Hambrick .1997. Explaining the premiums paid for large acquisitions: evidence of CEO hubris. Administrative Science Quarterly 42 (1): 103-127. Healy, P., K. Palepu and R. Ruback. 1992. Does corporate Performance Improve After Mergers? Journal of Financial Economics 31: 135-175. Hendry, J. 2003. Understanding Japanese Society. London, New York: RoutledgeCurzon. Heron R., and E. Lie. 2002. Operating Performance and the Method of Payment in Takeovers. Journal of Financial and Quantative Analysis 37 (1): 137-155. Hill, C.W.L. 1995. National Institutional Structures, Transaction Cost Economizing and Competitive Advantage: The Case of Japan. Organization Science 6 (1): 119-131. Hodder, J.E. and A.E. Tschoegl. 1993. “Corporate finance in Japan”. In S. Takagi, eds., Japanese capital markets. Cambridge: Blackwell Publishers. Holl, P. and D. Kyriazis. 1997. Wealth Creation and Bid Resistance in U.K. Takeover Bids. Strategic Management Journal 18 (6): 483-498. Hooker, R. 1996. Jomon & Yayoi.At http://www.wsu.edu:8000/~dee/ANCJAPAN/ANCJAPAN.HTM Horiuchi, A., F. Packer and S. Fukuda. 1988. What Role has the Main Bank Played in Japan? Journal of the Japanese and International Economies 2: 159-80. Horiuchi, T. 1994. The effect of firm status on banking relationships and loan syndication. In Masahiko Aoki and Hugh Patrick, eds., The Main Bank System. New York: Oxford University Press. Hoshi T. and A. Kashyap. 2001. Corporate Financing and Governance in Japan: The Road to the Future. MIT Press. Hoshi, T. 1995. “Evolution of the main bank system”. In Mitsuaki Okabe, ed., The structure of the Japanese Economy: Changes on the Domestic and International Fronts. New York: St. Martin’s Press.

228

240 References

Hoshi , T. 1998. “Japanese Corporate Governance as a System”. In K. Hopt. et al., eds. Comparative Corporate Governance: The state of the art and emerging research. Oxford: Oxford University Press. Hoshi, T., A. Kashap and D. Scharfstein. 1990. The role of banks in reducing the costs of financial distress in Japan. Journal of financial economics 27: 67-88. Hoshi, T., A. Kashap and D. Scharfstein. 1991. Corporate Structure, Liquidity and Investment: Evidence from Japanese industrial groups. Quaterly Journal of Economics 106: 33-60. Hoshino, Y. 1982. The performance of corporate mergers in Japan. Journal of Business Finance and Accounting 9: 153–165. Hoshino, Y. 1984. The Anatomy of Japanese business. New York: M.E. Sharpe, Inc. Hubbard, R.G. and D. Palia. 1999. A Reexamination of the Conglomerate Merger Wave in the 1960s: An Internal Capital Markets View. Journal of Finance 54, 1131-1152. Ikeda, K. and Doi, N. 1983. The Performances of Merging Firms in Japanese Manufacturing Industry: 1964-75. The Journal of Industrial Economics 31(3): 257-266. Ito, A. 1997. Aspects of Japanese culture and society : an anthropologist’s view. Tokyo : International Society for Educational Information Ito. 1989. M&A ni yoru Kigyǀ - Kigyo Kachi Souzou Kouka no Nichibei Hikaku (Company Value Creation through M&A – A Comparison between Japan and the US). Shoken Analyst Journal (98.7): 11-23. Jacobson, R. and D. Aaker. 1993. Myopic Management Behavior With Efficient, But Imperfect, Financial Markets: A Comparison Of Information Asymmetries In The U.S. And Japan. Journal of Accounting & Economics, 16, 383-405. Jensen, M. 1993. The modern industrial revolution, exit, and the failure of internal control systems. Journal of Finance 48: 831–80. Jensen, M. and R. Ruback. 1983. The Market for Corporate Control: Scientific Evidence. Journal of Financial Economics 11: 5-50. Jensen, M.C. 1986. Agency Costs of Free Cash Flow, Corporate Finance and Takeover. American Review 76: 323-329. Jensen, M.C. 1988. Takeover: Their causes and consequences. Journal of Economic Perspective 1: 21-48. Jensen, M.C. 1989. Eclipse of the public corporation. Harvard Business Review 5: 61-74. Jetro. 2001. Japan’s changing economic laws and regulations. Tokyo: Jetro Publication. Johnson, C. 1982. Miti and the Japanese Miracle The Growth of Industrial Policy, 1925- 1975. Stanford: Stanford University Press. Kaizuka, K. and E. Ono, eds. 1986. Nihon no kinyu shisutemu (The Japanese financial system). Tokyo: Tokyo University Press. Kang, J.K. and A. Shivdasani. 1995. Firm performance, corporate governance, and top executive turnover in Japan. Journal of financial economics 38: 29-58. Kang, J.K., A. Shivdasani and T. Yamada. 2000. The effect of bank relations on investment decisions: an investigation of Japanese takeover bids. Journal of Finance 55: 2197-2218. Kang, J-K. and R.M. Stulz. 1996. How Different is Japanese Corporate Finance? Review of Financial Studies 9(1): 109-139. Kaplan, S. and M. Weisbach. 1992. Evidence From Divestures. Journal of Finance 47: 107-138.

229

241 References

Kaplan, S.N. 1994. Top executive rewards and firm performance: A comparison of Japan and the United States. Journal of Political Economy 102: 510-546. Kaplan, S.N. and B.A. Minton. 1994. Appointments of outsiders to Japanese boards - Determinants and implications for managers. Journal of Financial Economics 36: 225- 258. Katz, J. 2002. Start Here: Social Ontology and Research. Strategy Theoretical Criminology 6(3): 255-278. Kawabe, N. 1989. The Development of Distribution Systems in Japan before World War II, Business and economic history, Second Series, Volume Eighteen: 33-44. Kester, C.W. 1986. Capital and Ownership Structure: a Comparison of United States and Japanese Manufacturing Corporations. Financial Management 15: 5-16. Kester, C.W. 1989. Industrial groups as systems of contractual governance. Oxford Review of Economic Policy 8 (3): 24-44. Kester, C.W. 1991. Japanese takeovers: the global contest for corporate control.Boston: Harvard Business School Press. Kitano, S. 1962. “Dozoku and Ie in Japan: The Meaning of Family Genealogical Relationships”. In R.J. Smith and R.K. Beardlsey, eds., Japanese Culture : Its Development and Characteristics. Klein, B. 1980. Transaction Cost Determinants of "Unfair" Contractual Arrangements. The American Economic Review 70 (2): 356-362. Kodate, H. 2005. The new Japanese corporation law in relation to M&A. The Asia Law Japan review 1(2):1-5. Komoto. 2002. Wagakuni no Kigyǀ Gappei no Tokutyo to Keizai Kouka (The Characteristics of Japanese Mergers and their Economic Effect). Kokumin Keizai Zasshi 186 (6): 1-16. Kosei Torihiki Iinkai. 1992. Nihon no Rokudai Kigyǀ Shǌdan (Six Largest Kigyǀ Shǌdan in Japan). Tokyo: Toyo Keizai Shinpo-sha. Kruse, T., H. Park, K. Park and K. Suzuki. 2007. Long-term performance following mergers of Japanese companies: The Effect of diversification and affiliation. Pacific- Basin Finance Journal 15 (2): 154-172. Kuroki, F. 2003. The Relationship of Companies and Banks as Cross-Shareholdings Unwind—Fiscal 2002 Cross-Shareholding Survey. NLI Research. Lang, L. and R.M. Stulz. 1994. Tobin’s q, Corporate Diversification and Firm Performance. Journal of Political Economy 102: 1248-1280. Lang, L., R.M. Stulz and R.A. Walkling. 1989. Managerial Performance, Tobin’s Q, and the Gains from Successful Tender Offers. Journal of Financial Economics 24 (1): 137- 154. Lang, L., R.M. Stulz and R.A. Walking. 1991. A test of the free cash flow hypothesis: The case of bidder returns. Journal of Financial Economics 29: 315-335. Langetieg, T.C. 1978. An Application of a Three-Factor Performance Index to Measure Stockholder Gains from Merger. Journal of Financial Economics 6: 365-383. LaPier, T. 1998. Competition, growth strategies, and the globalization of services : real estate advisory services in Japan, Europe and the United States. London: Routledge. Lebrun, L.N. 2001. Recent Amendments to the Commercial Code of Japan: Impact on Mergers and Acquisitions. Law & Policy in International Business 32: 811-823. Lin, B.X., D. Michayluk, H.R. Oppenheimer and S.F. Reid. 2008. Hubris amongst Japanese bidders. Pacific-Basin Finance Journal 16: 121–159.

230

242 References

Lin, C.C. 2005. The transition of the Japanese keiretsu in the changing economy. Journal of The Japanese and International Economies 19: 96-109. Loderer, M. and K. Martin. 1990. Corporate acquisitions by listed firms: the experience of a comprehensive sample. Financial Management 19, 17-33. Lynn, L.H. and T.J. McKeown. 1988. Organizing Business: Trade Associations in America and Japan. Washington: American Enterprise Institute for Public Policy Research. Malatesta P.H. 1983. The wealth effect of merger activity and the objective functions of merging firms. Journal of Financial Economics 11: 155-181. Malmendier, U. and G. Tate. 2003. Who makes acquisitions? CEO overconfidence and the market’s reaction. NBER working paper No.10813. Mandelker, G. 1974. Risk and return: The case of merging companies. Journal of Financial Economics 1: 303-335. Mason, M. 1997. Europe and the Japanese Challenge: The Regulation of Multinationals in Comparative Perspective.OxfordUniversity Matsusaka, J.G. 1993. Takeover Motives During the Conglomerate Merger Wave. RAND Journal of Economics 24: 357-379. Melicher, R.W. and D.F. Rush. 1974. Evidence on the Acquisition-Related Performance of Conglomerate Firms. The Journal of Finance 29 (1): 141-149. Merton, R.K. 1948. The self-fulfilling prophecy. Antioch Review 8: 193-210. Milhaupt, C. 2001. Creative Norm Destruction: The Evolution of Nonlegal Rules in Japanese Corporate Governance. University of Pennsylvania Law Review 149 (6) Milhaupt, C.J. and M.D. West. 2001. Institutional Change and M&A in Japan: Diversity Through Deals. Columbia Law School Working Paper No. 193. Milhaupt, C.J. and M.D. West. 2004. Economic Organizations and Corporate Governance in Japan: The Impact of Formal and Informal Rules. Oxford: Oxford University Press. Mitchell, M.L. and J.H. Mulherin. 1996. The Impact of Industry Shocks on Takeover and Restructuring Activity. Journal of Financial Economics 41: 193-229. Miwa, Y. and J.M. Ramseyer. 2001. The Fable of the Keiretsu. Harvard Law School, Discussion Paper No. 316. Miyajima H. and K. Fumiaki. 2005. The Unwinding of Cross-shareholding: Causes, Effects, and Implications. RIETI Discussion Paper Series 05-E-006. Research Institute of Economy, Trade and Industry (RIETI). Miyajima, H. 1994. The Transformation of prewar Zaibatsu to postwar Corporate Groups: From Hierarchical Integrated Group to Horizontally Integrated Group. The Journal of Japanese and International Economies (10). Miyajima, H. 2007. The Comparative Features and Economic Role of Mergers and Acquisitions in Japan. Unpublished working paper. Miyajima, H., T. Imajoh and S. Kawamoto. 2007. Mergers and Acquisitions in Japan from a Historical Perspective: Understanding M&A Waves in the Prewar Era. Unpublished paper. Morck, R. and M. Nakamura. 1999. Banks and Corporate Control in Japan. Journal of Finance 54: 319-339. Morck, R. and M. Nakamura. 2003. Been There, Done That The History of Corporate Ownership in Japan. ECGI Working Paper Series in Finance Working Paper N° 20/2003.

231

243 References

Morck, R., A. Shleifer and R. Vishny. 1990. Do Managerial Objectives Drive Bad Acquisitions? Journal of Finance 45 (1): 31-48. Morck, R., M. Nakamura and A. Shivdasani. 2000. Banks, ownership structure and firm value in Japan. Journal of Business 73 (4): 539-567. Moulton, H.G. 1931. Japan, an Economic and Financial Appraisal. Washington, D.C.: The Brookings Institution. Mueller, D.C. 1969. A Theory of Conglomerate Mergers, Quarterly Journal of Economics 83, 643-59. Mueller, D.C., ed. 1980. The Determinants and Effects of Mergers: An International Comparison.Cambridge,Mass.:Oelgeschlager,Gunn,andHain. Mueller, D.C. 2003. The Corporation : Investment, mergers, and growth. London, New York: Routledge. Muramatsu, S. 1986. Zaimu de-ta ni yoru gappei kouka no bunseki (Analyses if the effects of mergers with financial data). Kigyou kaikei 38: 668-677. Myers, S.C. 1984. The Capital Structure Puzzle. The Journal of Finance 39: 575-592. Nakamura, H. 1958. Some features of the Japanese way of thinking. Monumenta Nipponica 14 (No.3/4): 277-318. Nakamura,T.1981.The postwar Japanese economy: its development and structure. Tokyo: Univ. of Tokyo Press. Nakamura,T.1994.Lectures on modern japanese economic history 1926- 1994. Tokyo: LTCB International Library Foundation. Nakatani, I. 1984. The Economic Role of Financial Corporate Grouping. In M. Aoki, ed., The Economic Analysis of the Japanese Firm. North-Holland: Elsevier Science Publishers B.V.: 227-258. Nelson, R.L. 1959. Merger movements in American industry, 1895-1956. Princeton: Princeton University Press. Nishiguchi, T. 1994. Strategic Industrial Sourcing. New York: Oxford University Press North, D.C. 1990. Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press. North, D.C. 1994. Economic Performance Through Time. American Economic Review 84 (3): 359-68. North, D.C. 2005. Understanding the process of economic change. Princeton: Princeton Univresity Press. Odagiri, H. 1992. Growth through competition, competition through growth : strategic management and the economy in Japan. Oxford: Clarendon Press. Odagiri, H. and T. Hase. 1989. Are mergers and acquisitions going to be popular in Japan too? International Journal of Industrial Organization 7: 49-72. Ohno, K. 2006. The Economic Development of Japan: The Path Traveled by Japan as a Developing Country. GRIPS Development Forum: Tokyo. Oka, J. 1993. The Stock Market (I). In FAIR fact series: Japan’s financial markets. Japan: Foundation for Advanced Information and Research. Okazaki, T. 1999. Economic history of the holding company: Zaibatau and corporate governance (Mochikabu kaishano keizaishi: zaibatsuto kigyǀkigyo tochi) (in Japanese). Chikuma Shobo Tokyo. Okazaki, T. 1999a. “Corporate Governance”. In T. Okazaki and M. Okuno-Fujiwara, eds. The Japanese Economic System and its Historical Origins. Oxford: Oxford University

232

244 References

Press. Translation of the 1993 Gendai Nihon Keizai System no Genryǌ (The Origin of the Contemporary Japanese Economic System), Tokyo: Nihon Keizai Shimbun-sha. Okazaki, T. 2000. Role of Holding Companies in Prewar Japanese Economic Development: Rethinking Zaibatsu in Perspectives of Corporate Governance. CIRJE Discussion Paper Series CIRJE-F-74. Okazaki, T. 2001. The role of Holding Companies in Pre-war Japanese Economic Development: Rethinking Zaibatsu in Perspectives of Corporate Governance. Social Science Japan Journal 4: 243-268. Okazaki, T. 2006. Micro-aspects of Monetary Policy in Pre-war Japan: Lender of Last Resort and Selection of Banks. CIRJE Discussion Paper Series: CIRJE-F-398. Okazaki, T., M. Sawada and K. Wang. 2005. The Fall of “Organ Bank” Relationships During the wave of Bank Failures and Consolidations: Experience in Pre-war Japan. CIRJE Discussion Paper Series: CIRJE-F-379. Okumura, H. 2000. Corporate capitalism in Japan. Basingstoke: Macmillan, New York : St. Martin's. Pettway, H.R. and T. Yamada. 1986. Mergers in Japan and their Impacts upon Stockholder’s Wealth. Financial Management 15 (4): 43-52. Porter, M. 1992. Capital Disadvantage: America’s Failing Capital Investment System. Harvard Business Review September-October: 65-83. Powell, W.W. 1990. Neither Market nor Hierarchy: Network Forms of Organization. Research in Organizational Behavior 12: 295-336. Prowse, S. 1990. Institutional Investment Patterns and Corporate Financial Behavior in the United States and Japan. Journal of Financial Economics 27: 43-66. Raj, M. and M. Forsyth. 2003. Hubris Amongst U.K. Bidders and Losses to Shareholders. International Journal of Business 8 (1): 1-16. Rau, P.R. and T. Vermaelen. 1998. Glamour, value and the post-acquisition performance of acquiring firms. Journal of Financial Economics 49, 223-253. Ravenscraft, D. and F.M. Scherer. 1987. Mergers, Sell-Offs, and Economic Efficiency. Washington, DC: Brookings Institution. Rhodes-Kropf, M. and S. Viswanathan. 2004. Market Valuation and Merger Waves. Journal of Finance 59: 2685-2718. Röhl, W. 2005. History of law in Japan since 1868.Leiden:Brill. Roll, R. 1986. The Hubris Hypothesis of Corporate Takeovers. Journal of Business 59: 197-216. Rutherford, M. 1994. Institutions in Economics: The Old and the New Institutionalism. New York: Cambridge University Press. Sako, M. 1991. The role of 'trust' in Japanese Buyer-Supplier Relationships. Ricerche Economiche 45 (2/3): 449-74. Sansom, G.B. 1978. Japan A Short Cultural History. Stanford, California: Stanford University Press. Sato, K. and Y. Hoshino, eds. 1984. The anatomy of Japanese business. New York: M. E. SHARPE, INC. Schaede, U. 2000. Cooperative capitalism : self-regulation, trade associations and the antimonopoly law in Japan. Oxford: Oxford University Press. Schaede, U. 2006. Competition for Corporate Control: Institutional Investors, Investment Funds, and Hostile Takeovers in Japan. Center on Japanese Economy and Business: Working Paper Series No. 248.

233

245 References

Scher, M. 2001. Bank-firm Cross-shareholding in Japan:What is it, why does it matter, is it winding down? DESA Discussion Paper No. 15. Schwert, G. W. 2000. Hostility in Takeovers: In the Eyes of the Beholder? The Journal of Finance 55 (6): 2599-2640. Sekiguchi, H. 2003. The Patriarchal Family Paradigm in Eighth-Century Japan. In D. Ko, J.K. Haboush and J.R. Piggott, eds., Women and Confucian cultures in premodern China, Korea, and Japan. Berkeley: University of California Press. Servaes, H. 1991. Tobin’s Q and the gains from takeovers. Journal of Finance 46, 409-419. Sheard, P. 1989. The main bank system and corporate monitoring and control in Japan. Journal of Economic Behavior and Organization 11, 399-422. Sheard, P. 1991. The economics of interlocking shareholding in Japan. Ricerche Economiche Vol. XLV (2-3): 421-48. Sheldon, C.D. 1958. The Rise of the Merchant Class in Tokugawa Japan 1600-1868 : An introductory survey. New York: J.J. Augustin Incorporated Publisher. Sherif, M. and C.W. Sherif. 1953. Groups in Harmony and Tension : An Integration of Studies on Intergroup Relations. New York: Harper & Brothers, Publishers. Shiba, T. and M. Shimotani. 1997. Beyond the firm: Business Groups in International and Historical Perspective. New York: Oxford University Press. Shibagaki, K. 1968: Mitsui-Mitsubishi no hyakunen – Nihon shihonshugi to zaibatsu (100 years Mitsui und Mitsubishi – The Japanese capitalism and zaibatsu). Tokyo: Chǌǀ Kǀronsha. Shishido, Z. 2004. Changes in Japanese Corporate Law and Governance: Revisiting the Convergence Debate. Law and Economics Workshop, Law paper No. 216. Shleifer, A. and R.W. Vishny. 2003. Stock Market Driven Acquisitions. Journal of Financial Economics 70: 295-489. Smitka, M.J. 1991. Competitive Ties: Subcontracting in the Japanese Automotive Industry. New York: Columbia University Press. Steenbeek, O.W., 1996. Financial Regulation in Japan; systemic risk and the Nikkei futures market (Ph.D. thesis). Amsterdam: Thesis Publishers. Stigler, G. 1950. Monopoly and Oligopoly Power by Merger. American Economic Review 40: 23-34. Stulz, R. 1990. Managerial discretion and optimal financing policies. Journal of Financial Economics 26: 3-27. Suzuki, K. 1997. From Zaibatsu to Corporate Complexes. In T. Shiba and M. Shimotani, eds., Beyond the firm: Business Groups in International and Historical Perspective. New York: Oxford University Press. Suzuki, K. 2002. Competition Law Reform in Britain and Japan: Comparative Analysis of Policy Network. London, New York: Routledge. Suzuki, T. 1996. Narrating the Self: Fictions of Japanese Modernity. Stanford, California: Stanford Unitversity Press. Tasugi, K. 1963. Business-chain between large and small firms. The Kyoto University Review 33 (2): 21-58. Tatsuta, M. 2005. Ongoing Modernization of Japanese Company Law. In K.J. Hopt et al. eds., Corporate governance in context : corporations, states and markets in Europe, Japan, and the US. Oxford: Oxford University Press.

234

246 References

Teranishi, J. 1994. Loan Syndication in War-Time Japan and the Origins of the Main Bank System. In Masahiko Aoki and Hugh Patrick, eds., The Main Bank System.NewYork: Oxford University Press. Teranishi, J. 2005. Evolution of the Economic System in Japan. Cheltenham: Edward Elgar. Travlos, N.G. 1987. Corporate Takeover Bids, Methods of Payment, and Bidding Firms’ Stock Returns. The Journal of Finance 42: 943-963. Wagatsuma, S. 1956. The Japanese Legal System 1945-1955: Changes in Retrospect. Monumenta Nipponica 12 (1/2): 105-119. Warner J.B. 1977. Bankruptcy Costs: Some Evidence. The Journal of Finance 32 (2): 337- 347. Watt, P. 2003. Japanese Religions. Japan Digest of October 2003. “http://iis- db.stanford.edu/docs/127/religion.pdf” Westney D.E. 2001. Multinational enterprises and cross-border knowledge creation. In I. Nonaka and T. Nishiguchi, eds., Knowledge Emergence. Oxford: Oxford University Weston, J. F. and S.K. Mansinghka. 1971. Tests of the Efficiency Performance of Conglomerate Firms. The Journal of Finance 26 (4): 919-936. Williamson , O.E. 1996. The Mechanisms of Governance. New York: Oxford University Press. Williamson, O.E. 1975. Markets and Hierarchies. Analysis and Antitrust Implications. NewYork:TheFreePress. Williamson, O.E. 1985. The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting. New York: Free Press. Williamson, O.E. 1991. Comparative Economic Organization: the Analysis of Discrete Structural Alternatives. Administrative Science Quarterly 36: 269-296. Williamson, O.E. 1998. Transaction Cost Economics: How it works; Where it is headed, De Economist 146 (1): 23-55. Yamagishi, T. and M. Yamagishi. 1998. “Trust and Commitment as Alternative Responses to Social Uncertainty”. In W.M. Fruin, ed., Networks, Markets, and the Pacific Rim: Studies in Strategy. New York: Oxford University Press. Yao, X. 2000. An introduction to Confucianism. Cambridge: Cambridge University Press. Yeh, T. 2007. Bank Directorships and Bidder’s Returns in Japanese Takeovers Japanese Journal of Administrative Science Article Volume 20, No.2, 2007, 169᧩184. Yeh, T., Y. Hoshino. 2002. Shareholders' Wealth, Bank Control, and Large Shareholders: An Analysis of Japanese Mergers Working Paper Series Vol. 2002-08. May 2002. Yeh, T. and Y. Hoshino. 2002a. Productivity and operating performance of Japanese merging firms: Keiretsu-related and independent mergers. Japan and the World Economy 14: 347–366.

235

247 References

OTHER REFERENCES:

Newspapers, various issues of: Nihon Keizai Shimbun (Japan Economic Journal), Nikkei Sangyǀ Shimbun (Industrial Journal), Nikkei Ryǌtǌ Shimbun (Distribution Journal), Nikkei Kinyǌ Shimbun (Finance Journal), The New York Times, Forbes, Bloomberg News, Reuters, Financial Times, The Asahi Shimbun, International Herald Tribune, Kyodo News International, South China Morning Post, Nikkei Business, and Wall Street Journal.

Statistical handbooks Bank of Japan. Economic Statistics of Japan. various issues. Bank of Japan. Flow of Funds. Various years. International Financial Statistics Yearbook. Various issues. International Monertary Fund, ed. JFTC Annual Reports. Various issues. Japan Fair Trade Commission, eds. Kabushiki Gaisha Nenkan (Joint-stock company Yearbook). 1937. Toyo Keizai Shinpo Sha, ed. Kigyǀ gurupu to gyoukai chizu (Company groups and industry overview). 2004. Tokyo: Takahashi shoten. Kigyǀ Keiretsu Soran, (Survey of Corporate Groups). Various issues. Toyo Keizai Shinpo Sha, ed. Ministry of Finance, Flow of Funds Tables (no distinction made between Stock and Bonds for period 1962-69). Nikkei Kaisha Jǀhǀ (Nikkei Company Information). Various issues. Nihon keizai shimbun sha, ed. Nikkei Kaisha Sokan (Nikkei; Annual Corporation Reports (Unlisted)). Various issues. Nihon keizai shimbun sha, ed. Nippon Kigyǀ Shǌdan for the years 1988 and 1998. Nomura Securities. ed. Various issues. Shouken toukei youran (Securities Statistics). Toyo Keizai Shinpo Sha. Recof Corporation. 2003, 2007. MARR Magazine: Tokyo.

Website Ministry of Finance Japan, Policy research institute “http://www.mof.go.jp/1c002.htm”

236

248 About the Author

Dimitri van Schaik was born in 1975 in Sliedrecht, the Netherlands. After receiving his Gymnasium’s diploma from the Bonifacius College in Utrecht, he started studying Japanese language and culture at the University of Leiden in 1994. In his first year he qualified for a one-year scholarship from the University of Leiden that enabled him to continue his study in Japan. While in Japan he decided to start studying Economics as well upon his return to the Netherlands in 1996. In 2000 he received both his MA degree in Japanese language and culture and his MSc degree in Business Economics. After graduation he worked for Mizuho Corporate Bank in the Netherlands. As an account officer, his main responsibilities were financial analysis of corporate clients and building good relationships to enhance the business. In 2003 he started his PhD project, after receiving a Monbukagakusho scholarship that enabled him to go to Hitotsubashi University in Tokyo as a research student. He conducted his research from 2003 to 2007 in Japan and then returned to the Netherlands for the finalization of his thesis. Currently, he is employed by PricewaterhouseCoopers in the Netherlands.

237

249 238

250 ERASMUS RESEARCH INSTITUTE OF MANAGEMENT (ERIM)

ERIM PH.D. SERIES RESEARCH IN MANAGEMENT

ERIM Electronic Series Portal: http://hdl.handle.net/1765/1

Althuizen, N.A.P., Analogical Reasoning as a Decision Support Principle for Weakly Structured Marketing Problems, Promotor: Prof. dr. ir. B. Wierenga, EPS-2006-095- MKT, ISBN: 90-5892-129-8, http://hdl.handle.net/1765/8190

Alvarez, H.L., Distributed Collaborative Learning Communities Enabled by Information Communication Technology, Promotor: Prof. dr. K. Kumar, EPS-2006-080-LIS, ISBN: 90-5892-112-3, http://hdl.handle.net/1765/7830

Appelman, J.H., Governance of Global Interorganizational Tourism Networks: Changing Forms of Co-ordination between the Travel Agency and Aviation Sector, Promotors: Prof. dr. F.M. Go & Prof. dr. B. Nooteboom, EPS-2004-036-MKT, ISBN: 90-5892-060-7, http://hdl.handle.net/1765/1199

Baquero, G, On Hedge Fund Performance, Capital Flows and Investor Psychology, Promotor: Prof. dr. M.J.C.M. Verbeek, EPS-2006-094-F&A, ISBN: 90-5892-131-X, http://hdl.handle.net/1765/8192

Berens, G., Corporate Branding: The Development of Corporate Associations and their Influence on Stakeholder Reactions, Promotor: Prof. dr. C.B.M. van Riel, EPS-2004-039- ORG, ISBN: 90-5892-065-8, http://hdl.handle.net/1765/1273

Berghe, D.A.F. van den, Working Across Borders: Multinational Enterprises and the Internationalization of Employment, Promotors: Prof. dr. R.J.M. van Tulder & Prof. dr. E.J.J. Schenk, EPS-2003-029-ORG, ISBN: 90-5892-05-34, http://hdl.handle.net/1765/1041

Berghman, L.A., Strategic Innovation Capacity: A Mixed Method Study on Deliberate Strategic Learning Mechanisms, Promotor: Prof. dr. P. Mattyssens, EPS-2006-087-MKT, ISBN: 90-5892-120-4, http://hdl.handle.net/1765/7991

239

251 ERIM PhD Series

Bijman, W.J.J., Essays on Agricultural Co-operatives: Governance Structure in Fruit and Vegetable Chains, Promotor: Prof. dr. G.W.J. Hendrikse, EPS-2002-015-ORG, ISBN: 90-5892-024-0, http://hdl.handle.net/1765/867

Bispo, A., Labour Market Segmentation: An investigation into the Dutch hospitality industry, Promotors: Prof. dr. G.H.M. Evers & Prof. dr. A.R. Thurik, EPS-2007-108- ORG, ISBN: 90-5892-136-9, http://hdl.handle.net/1765/10283

Blindenbach-Driessen, F., Innovation Management in Project-Based Firms, Promotor: Prof. dr. S.L. van de Velde, EPS-2006-082-LIS, ISBN: 90-5892-110-7, http://hdl.handle.net/1765/7828

Boer, C.A., Distributed Simulation in Industry, Promotors: Prof. dr. A. de Bruin & Prof. dr. ir. A. Verbraeck, EPS-2005-065-LIS, ISBN: 90-5892-093-3, http://hdl.handle.net/1765/6925

Boer, N.I., Knowledge Sharing within Organizations: A situated and Relational Perspective, Promotor: Prof. dr. K. Kumar, EPS-2005-060-LIS, ISBN: 90-5892-086-0, http://hdl.handle.net/1765/6770

Boer-Sorbán, K., Agent-Based Simulation of Financial Markets: A modular, Continuous- Time Approach, Promotor: Prof. dr. A. de Bruin, EPS-2008-119-LIS, ISBN: 90-5892- 155-0, http://hdl.handle.net/1765/10870

Boon, C.T., HRM and Fit: Survival of the Fittest!?, Promotors: Prof. dr. J. Paauwe & Prof. dr. D.N. den Hartog, EPS-2008-129-ORG, ISBN: 978-90-5892-162-8, http://hdl.handle.net/1765/1

Brito, M.P. de, Managing Reverse Logistics or Reversing Logistics Management? Promotors: Prof. dr. ir. R. Dekker & Prof. dr. M. B. M. de Koster, EPS-2004-035-LIS, ISBN: 90-5892-058-5, http://hdl.handle.net/1765/1132

Brohm, R., Polycentric Order in Organizations: A Dialogue between Michael Polanyi and IT-Consultants on Knowledge, Morality, and Organization, Promotors: Prof. dr. G. W. J. Hendrikse & Prof. dr. H. K. Letiche, EPS-2005-063-ORG, ISBN: 90-5892-095-X, http://hdl.handle.net/1765/6911

Brumme, W.-H., Manufacturing Capability Switching in the High-Tech Electronics Technology Life Cycle, Promotors:Prof.dr.ir.J.A.E.E.vanNunen&Prof.dr.ir.L.N. Van Wassenhove, EPS-2008-126-LIS, ISBN: 978-90-5892-150-5, http://hdl.handle.net/1765/1

Campbell, R.A.J., Rethinking Risk in International Financial Markets, Promotor: Prof. dr. C.G. Koedijk, EPS-2001-005-F&A, ISBN: 90-5892-008-9, http://hdl.handle.net/1765/306

240

252 ERIM PhD Series

Chen, H., Individual Mobile Communication Services and Tariffs, Promotor: Prof. dr. L.F.J.M. Pau, EPS-2008-123-LIS, ISBN: 90-5892-158-1, http://hdl.handle.net/1765/11141

Chen, Y., Labour Flexibility in China’s Companies: An Empirical Study, Promotors: Prof. dr. A. Buitendam & Prof. dr. B. Krug, EPS-2001-006-ORG, ISBN: 90-5892-012-7, http://hdl.handle.net/1765/307

Damen, F.J.A., Taking the Lead: The Role of Affect in Leadership Effectiveness, Promotor: Prof. dr. D.L. van Knippenberg, EPS-2007-107-ORG, http://hdl.handle.net/1765/10282

Daniãevská, P., Empirical Studies on Financial Intermediation and Corporate Policies, Promotor: Prof. dr. C.G. Koedijk, EPS-2004-044-F&A, ISBN: 90-5892-070-4, http://hdl.handle.net/1765/1518

Delporte-Vermeiren, D.J.E., Improving the Flexibility and Profitability of ICT-enabled Business Networks: An Assessment Method and Tool, Promotors: Prof. mr. dr. P.H.M. Vervest & Prof. dr. ir. H.W.G.M. van Heck, EPS-2003-020-LIS, ISBN: 90-5892-040-2, http://hdl.handle.net/1765/359

Derwall, J.M.M., The Economic Virtues of SRI and CSR, Promotor: Prof. dr. C.G. Koedijk, EPS-2007-101-F&A, ISBN: 90-5892-132-8, http://hdl.handle.net/1765/8986

Dijksterhuis, M., Organizational Dynamics of Cognition and Action in the Changing Dutch and US Banking Industries, Promotors: Prof. dr. ir. F.A.J. van den Bosch & Prof. dr. H.W. Volberda, EPS-2003-026-STR, ISBN: 90-5892-048-8, http://hdl.handle.net/1765/1037

Elstak, M.N., Flipping the Identity Coin: The Comparative Effect of Perceived, Projected and Desired Organizational Identity on Organizational Identification and Desired Behavior, Promotor: Prof. dr. C.B.M. van Riel, EPS-2008-117-ORG, ISBN: 90-5892- 148-2, http://hdl.handle.net/1765/10723

Fenema,P.C.van,Coordination and Control of Globally Distributed Software Projects, Promotor: Prof. dr. K. Kumar, EPS-2002-019-LIS, ISBN: 90-5892-030-5, http://hdl.handle.net/1765/360

Fleischmann, M., Quantitative Models for Reverse Logistics, Promotors: Prof. dr. ir. J.A.E.E. van Nunen & Prof. dr. ir. R. Dekker, EPS-2000-002-LIS, ISBN: 35-4041-711-7, http://hdl.handle.net/1765/1044

Flier, B., Strategic Renewal of European Financial Incumbents: Coevolution of Environmental Selection, Institutional Effects, and Managerial Intentionality, Promotors: Prof. dr. ir. F.A.J. van den Bosch & Prof. dr. H.W. Volberda, EPS-2003-033-STR, ISBN: 90-5892-055-0, http://hdl.handle.net/1765/1071

241

253 ERIM PhD Series

Fok, D., Advanced Econometric Marketing Models, Promotor: Prof. dr. Ph.H.B.F. Franses, EPS-2003-027-MKT, ISBN: 90-5892-049-6, http://hdl.handle.net/1765/1035

Ganzaroli, A., Creating Trust between Local and Global Systems, Promotors: Prof. dr. K. Kumar & Prof. dr. R.M. Lee, EPS-2002-018-LIS, ISBN: 90-5892-031-3, http://hdl.handle.net/1765/361

Gilsing, V.A., Exploration, Exploitation and Co-evolution in Innovation Networks, Promotors: Prof. dr. B. Nooteboom & Prof. dr. J.P.M. Groenewegen, EPS-2003-032- ORG, ISBN: 90-5892-054-2, http://hdl.handle.net/1765/1040

Ginkel, W.P. van, The Use of Distributed Information in Decision Making Groups: The Role of Shared Task Representations, Promotor: Prof. dr. D. van Knippenberg, EPS- 2007-097-ORG, http://hdl.handle.net/1765/8424

Govers, R., Virtual Tourism Destination Image: Glocal Identities Constructed, Perceived and Experienced, Promotors: Prof. dr. F.M. Go & Prof. dr. K. Kumar, EPS-2005-069- MKT, ISBN: 90-5892-107-7, http://hdl.handle.net/1765/6981

Graaf, G. de, Tractable Morality: Customer Discourses of Bankers, Veterinarians and Charity Workers, Promotors: Prof. dr. F. Leijnse & Prof. dr. T. van Willigenburg, EPS- 2003-031-ORG, ISBN: 90-5892-051-8, http://hdl.handle.net/1765/1038

Groot, E.A. de, Essays on Economic Cycles, Promotors: Prof. dr. Ph.H.B.F. Franses & Prof. dr. H.R. Commandeur, EPS-2006-091-MKT, ISBN: 90-5892-123-9, http://hdl.handle.net/1765/8216

Gutkowska, A.B., Essays on the Dynamic Portfolio Choice, Promotor: Prof. dr. A.C.F. Vorst, EPS-2006-085-F&A, ISBN: 90-5892-118-2, http://hdl.handle.net/1765/7994

Hagemeijer, R.E., The Unmasking of the Other, Promotors: Prof. dr. S.J. Magala & Prof. dr. H.K. Letiche, EPS-2005-068-ORG, ISBN: 90-5892-097-6, http://hdl.handle.net/1765/6963

Halderen, M.D. van, Organizational Identity Expressiveness and Perception Management: Principles for Expressing the Organizational Identity in Order to Manage the Perceptions and Behavioral Reactions of External Stakeholders, Promotor: Prof. dr. S.B.M. van Riel, EPS-2008-122-ORG, ISBN: 90-5892-153-6, http://hdl.handle.net/1765/10872

Hartigh, E. den, Increasing Returns and Firm Performance: An Empirical Study, Promotor: Prof. dr. H.R. Commandeur, EPS-2005-067-STR, ISBN: 90-5892-098-4, http://hdl.handle.net/1765/6939

242

254 ERIM PhD Series

Hermans. J.M., ICT in Information Services; Use and Deployment of the Dutch Securities Trade, 1860-1970, Promotor: Prof. dr. drs. F.H.A. Janszen, EPS-2004-046-ORG, ISBN 90-5892-072-0, http://hdl.handle.net/1765/1793

Heugens, P.P.M.A.R., Strategic Issues Management: Implications for Corporate Performance, Promotors: Prof. dr. ir. F.A.J. van den Bosch & Prof. dr. C.B.M. van Riel, EPS-2001-007-STR, ISBN: 90-5892-009-7, http://hdl.handle.net/1765/358

Heuvel, W. van den, The Economic Lot-Sizing Problem: New Results and Extensions, Promotor: Prof. dr. A.P.L. Wagelmans, EPS-2006-093-LIS, ISBN: 90-5892-124-7, http://hdl.handle.net/1765/1805

Hoedemaekers, C.M.W., Performance, Pinned down: A Lacanian Analysis of Subjectivity at Work, Promotors: Prof. dr. S. Magala & Prof. dr. D.H. den Hartog, EPS-2008-121- ORG, ISBN: 90-5892-156-7, http://hdl.handle.net/1765/10871

Hooghiemstra, R., The Construction of Reality: Cultural Differences in Self-serving Behaviour in Accounting Narratives, Promotors: Prof. dr. L.G. van der Tas RA & Prof. dr. A.Th.H. Pruyn, EPS-2003-025-F&A, ISBN: 90-5892-047-X, http://hdl.handle.net/1765/871

Hu, Y., Essays on the Governance of Agricultural Products: Cooperatives and Contract Farming, Promotors: Prof. dr. G.W.J. Hendrkse & Prof. Dr. B. Krug, EPS-2007-113- ORG, ISBN: 90-5892-145-1, http://hdl.handle.net/1765/10535

Huij, J.J., New Insights into Mutual Funds: Performance and Family Strategies, Promotor: Prof. dr. M.C.J.M. Verbeek, EPS-2007-099-F&A, ISBN: 90-5892-134-4, http://hdl.handle.net/1765/9398

Huurman, C.I., Dealing with Electricity Prices, Promotor: Prof. dr. C.D. Koedijk, EPS- 2007-098-F&A, ISBN: 90-5892-130-1, http://hdl.handle.net/1765/9399

Iastrebova, K, Manager’s Information Overload: The Impact of Coping Strategies on Decision-Making Performance, Promotor: Prof. dr. H.G. van Dissel, EPS-2006-077-LIS, ISBN: 90-5892-111-5, http://hdl.handle.net/1765/7329

Iwaarden, J.D. van, Changing Quality Controls: The Effects of Increasing Product Variety and Shortening Product Life Cycles, Promotors: Prof. dr. B.G. Dale & Prof. dr. A.R.T. Williams, EPS-2006-084-ORG, ISBN: 90-5892-117-4, http://hdl.handle.net/1765/7992

Jansen, J.J.P., Ambidextrous Organizations, Promotors: Prof. dr. ir. F.A.J. Van den Bosch & Prof. dr. H.W. Volberda, EPS-2005-055-STR, ISBN: 90-5892-081-X, http://hdl.handle.net/1765/6774

243

255 ERIM PhD Series

Jong, C. de, Dealing with Derivatives: Studies on the Role, Informational Content and Pricing of Financial Derivatives, Promotor: Prof. dr. C.G. Koedijk, EPS-2003-023-F&A, ISBN: 90-5892-043-7, http://hdl.handle.net/1765/1043

Keizer, A.B., The Changing Logic of Japanese Employment Practices: A Firm-Level Analysis of Four Industries, Promotors: Prof. dr. J.A. Stam & Prof. dr. J.P.M. Groenewegen, EPS-2005-057-ORG, ISBN: 90-5892-087-9, http://hdl.handle.net/1765/6667

Kijkuit, R.C., Social Networks in the Front End: The Organizational Life of an Idea, Promotor: Prof. dr. B. Nooteboom, EPS-2007-104-ORG, ISBN: 90-5892-137-6, http://hdl.handle.net/1765/10074

Kippers, J., Empirical Studies on Cash Payments, Promotor: Prof. dr. Ph.H.B.F. Franses, EPS-2004-043-F&A, ISBN: 90-5892-069-0, http://hdl.handle.net/1765/1520

Kole, E., On Crises, Crashes and Comovements, Promotors: Prof. dr. C.G. Koedijk & Prof. dr. M.J.C.M. Verbeek, EPS-2006-083-F&A, ISBN: 90-5892-114-X, http://hdl.handle.net/1765/7829

Kooij-de Bode, J.M., Distributed Information and Group Decision-Making: Effects of Diversity and Affect, Promotor: Prof. dr. D.L. van Knippenberg, EPS-2007-115-ORG, http://hdl.handle.net/1765/10722

Knapp, S., The Econometrics of Maritime Safety: Recommendations to Enhance Safety at Sea, Promotor: Prof. dr. Ph.H.B.F. Franses, EPS-2007-096-ORG, ISBN: 90-5892-127-1, http://hdl.handle.net/1765/7913

Koppius, O.R., Information Architecture and Electronic Market Performance, Promotors: Prof. dr. P.H.M. Vervest & Prof. dr. ir. H.W.G.M. van Heck, EPS-2002-013-LIS, ISBN: 90-5892-023-2, http://hdl.handle.net/1765/921

Kotlarsky, J., Management of Globally Distributed Component-Based Software Development Projects, Promotor: Prof. dr. K. Kumar, EPS-2005-059-LIS, ISBN: 90- 5892-088-7, http://hdl.handle.net/1765/6772

Kuilman, J., The Re-Emergence of Foreign Banks in Shanghai: An Ecological Analysis, Promotor: Prof. dr. B. Krug, EPS-2005-066-ORG, ISBN: 90-5892-096-8, http://hdl.handle.net/1765/6926

Langen, P.W. de, The Performance of Seaport Clusters: A Framework to Analyze Cluster Performance and an Application to the Seaport Clusters of Durban, Rotterdam and the Lower Mississippi, Promotors: Prof. dr. B. Nooteboom & Prof. drs. H.W.H. Welters, EPS-2004-034-LIS, ISBN: 90-5892-056-9, http://hdl.handle.net/1765/1133

244

256 ERIM PhD Series

Le Anh, T., Intelligent Control of Vehicle-Based Internal Transport Systems, Promotors: Prof. dr. M.B.M. de Koster & Prof. dr. ir. R. Dekker, EPS-2005-051-LIS, ISBN: 90- 5892-079-8, http://hdl.handle.net/1765/6554

Le-Duc, T., Design and Control of Efficient Order Picking Processes, Promotor: Prof. dr. M.B.M. de Koster, EPS-2005-064-LIS, ISBN: 90-5892-094-1, http://hdl.handle.net/1765/6910

Leeuwen, E.P. van, Recovered-Resource Dependent Industries and the Strategic Renewal of Incumbent Firm: A Multi-Level Study of Recovered Resource Dependence Management and Strategic Renewal in the European Paper and Board Industry, Promotors: Prof. dr. ir. F.A.J. Van den Bosch & Prof. dr. H.W. Volberda, EPS-2007-109- STR, ISBN: 90-5892-140-6, http://hdl.handle.net/1765/10183

Lentink, R.M., Algorithmic Decision Support for Shunt Planning, Promotors: Prof. dr. L.G. Kroon & Prof. dr. ir. J.A.E.E. van Nunen, EPS-2006-073-LIS, ISBN: 90-5892-104- 2, http://hdl.handle.net/1765/7328

Liang, G., New Competition: Foreign Direct Investment and Industrial Development in China, Promotor: Prof. dr. R.J.M. van Tulder, EPS-2004-047-ORG, ISBN: 90-5892-073- 9, http://hdl.handle.net/1765/1795

Liere, D.W. van, Network Horizon and the Dynamics of Network Positions: A Multi- Method Multi-Level Longitudinal Study of Interfirm Networks, Promotor: Prof. dr. P.H.M. Vervest, EPS-2007-105-LIS, ISBN: 90-5892-139-0, http://hdl.handle.net/1765/10181

Loef, J., Incongruity between Ads and Consumer Expectations of Advertising,Promotors: Prof. dr. W.F. van Raaij & Prof. dr. G. Antonides, EPS-2002-017-MKT, ISBN: 90-5892- 028-3, http://hdl.handle.net/1765/869

Maeseneire, W., de, Essays on Firm Valuation and Value Appropriation, Promotor: Prof. dr. J.T.J. Smit, EPS-2005-053-F&A, ISBN: 90-5892-082-8, http://hdl.handle.net/1765/6768

Londoño, M. del Pilar, Institutional Arrangements that Affect Free Trade Agreements: Economic Rationality Versus Interest Groups, Promotors: Prof. dr. H.E. Haralambides & Prof. dr. J.F. Francois, EPS-2006-078-LIS, ISBN: 90-5892-108-5, http://hdl.handle.net/1765/7578

Maas, A.A., van der, Strategy Implementation in a Small Island Context: An Integrative Framework, Promotor: Prof. dr. H.G. van Dissel, EPS-2008-127-LIS, ISBN: 978-90- 5892-160-4, http://hdl.handle.net/1765/1

245

257 ERIM PhD Series

Maeseneire, W., de, Essays on Firm Valuation and Value Appropriation, Promotor: Prof. dr. J.T.J. Smit, EPS-2005-053-F&A, ISBN: 90-5892-082-8, http://hdl.handle.net/1765/6768

Mandele, L.M., van der, Leadership and the Inflection Point: A Longitudinal Perspective, Promotors: Prof. dr. H.W. Volberda & Prof. dr. H.R. Commandeur, EPS-2004-042-STR, ISBN: 90-5892-067-4, http://hdl.handle.net/1765/1302

Meer, J.R. van der, Operational Control of Internal Transport, Promotors: Prof. dr. M.B.M. de Koster & Prof. dr. ir. R. Dekker, EPS-2000-001-LIS, ISBN: 90-5892-004-6, http://hdl.handle.net/1765/859

Mentink, A., Essays on Corporate Bonds, Promotor: Prof. dr. A.C.F. Vorst, EPS-2005- 070-F&A, ISBN: 90-5892-100-X, http://hdl.handle.net/1765/7121

Meyer, R.J.H., Mapping the Mind of the Strategist: A Quantitative Methodology for Measuring the Strategic Beliefs of Executives, Promotor: Prof. dr. R.J.M. van Tulder, EPS-2007-106-ORG, ISBN: 978-90-5892-141-3, http://hdl.handle.net/1765/10182

Miltenburg, P.R., Effects of Modular Sourcing on Manufacturing Flexibility in the Automotive Industry: A Study among German OEMs, Promotors: Prof. dr. J. Paauwe & Prof. dr. H.R. Commandeur, EPS-2003-030-ORG, ISBN: 90-5892-052-6, http://hdl.handle.net/1765/1039

Moerman, G.A., Empirical Studies on Asset Pricing and Banking in the Euro Area, Promotor: Prof. dr. C.G. Koedijk, EPS-2005-058-F&A, ISBN: 90-5892-090-9, http://hdl.handle.net/1765/6666

Mol, M.M., Outsourcing, Supplier-relations and Internationalisation: Global Source Strategy as a Chinese Puzzle, Promotor: Prof. dr. R.J.M. van Tulder, EPS-2001-010- ORG, ISBN: 90-5892-014-3, http://hdl.handle.net/1765/355

Mom, T.J.M., Managers’ Exploration and Exploitation Activities: The Influence of Organizational Factors and Knowledge Inflows, Promotors: Prof. dr. ir. F.A.J. Van den Bosch & Prof. dr. H.W. Volberda, EPS-2006-079-STR, ISBN: 90-5892-116-6, http://hdl.handle.net/1765

Mulder, A., Government Dilemmas in the Private Provision of Public Goods, Promotor: Prof. dr. R.J.M. van Tulder, EPS-2004-045-ORG, ISBN: 90-5892-071-2, http://hdl.handle.net/1765/1790

Muller, A.R., The Rise of Regionalism: Core Company Strategies Under The Second Wave of Integration, Promotor: Prof. dr. R.J.M. van Tulder, EPS-2004-038-ORG, ISBN: 90-5892-062-3, http://hdl.handle.net/1765/1272

246

258 ERIM PhD Series

Ning, H., Hierarchical Portfolio Management: Theory and Applications, Promotor: Prof. dr. J. Spronk, EPS-2007-118-F&A, ISBN: 90-5892-152-9, http://hdl.handle.net/1765/10868

Noeverman, J., Management Control Systems, Evaluative Style, and Behaviour: Exploring the Concept and Behavioural Consequences of Evaluative Style, Promotors: Prof. dr. E.G.J. Vosselman & Prof. dr. A.R.T. Williams, EPS-2007-120-F&A, ISBN: 90- 5892-151-2, http://hdl.handle.net/1765/10869

Oosterhout, J., van, The Quest for Legitimacy: On Authority and Responsibility in Governance, Promotors: Prof. dr. T. van Willigenburg & Prof.mr. H.R. van Gunsteren, EPS-2002-012-ORG, ISBN: 90-5892-022-4, http://hdl.handle.net/1765/362

Paape, L., Corporate Governance: The Impact on the Role, Position, and Scope of Services of the Internal Audit Function, Promotors: Prof. dr. G.J. van der Pijl & Prof. dr. H. Commandeur, EPS-2007-111-MKT, ISBN: 90-5892-143-7, http://hdl.handle.net/1765/10417

Pak, K., Revenue Management: New Features and Models, Promotor: Prof. dr. ir. R. Dekker, EPS-2005-061-LIS, ISBN: 90-5892-092-5, http://hdl.handle.net/1765/362/6771

Pattikawa, L.H, Innovation in the Pharmaceutical Industry: Evidence from Drug Introduction in the U.S., Promotors: Prof. dr. H.R.Commandeur, EPS-2007-102-MKT, ISBN: 90-5892-135-2, http://hdl.handle.net/1765/9626

Peeters, L.W.P., Cyclic Railway Timetable Optimization, Promotors: Prof. dr. L.G. Kroon & Prof. dr. ir. J.A.E.E. van Nunen, EPS-2003-022-LIS, ISBN: 90-5892-042-9, http://hdl.handle.net/1765/429

Pietersz, R., Pricing Models for Bermudan-style Interest Rate Derivatives, Promotors: Prof. dr. A.A.J. Pelsser & Prof. dr. A.C.F. Vorst, EPS-2005-071-F&A, ISBN: 90-5892- 099-2, http://hdl.handle.net/1765/7122

Popova, V., Knowledge Discovery and Monotonicity, Promotor: Prof. dr. A. de Bruin, EPS-2004-037-LIS, ISBN: 90-5892-061-5, http://hdl.handle.net/1765/1201

Pouchkarev, I., Performance Evaluation of Constrained Portfolios, Promotors: Prof. dr. J. Spronk & Dr. W.G.P.M. Hallerbach, EPS-2005-052-F&A, ISBN: 90-5892-083-6, http://hdl.handle.net/1765/6731

Prins, R., Modeling Consumer Adoption and Usage of Value-Added Mobile Services, Promotors: Prof. dr. Ph.H.B.F. Franses & Prof. dr. P.C. Verhoef, EPS-2008-128-MKT, ISBN: 978/90-5892-161-1, http://hdl.handle.net/1765/1

247

259 ERIM PhD Series

Puvanasvari Ratnasingam, P., Interorganizational Trust in Business to Business E- Commerce, Promotors: Prof. dr. K. Kumar & Prof. dr. H.G. van Dissel, EPS-2001-009- LIS, ISBN: 90-5892-017-8, http://hdl.handle.net/1765/356

Quak, H.J., Sustainability of Urban Freight Transport: Retail Distribution and Local Regulation in Cities, Promotor: Prof. dr.M.B.M. de Koster, EPS-2008-124-LIS, ISBN: 978-90-5892-154-3, http://hdl.handle.net/1765/11990

Rinsum,M.van,Performance Measurement and Managerial Time Orientation, Promotor: Prof. dr. F.G.H. Hartmann, EPS-2006-088-F&A, ISBN: 90-5892-121-2, http://hdl.handle.net/1765/7993

Romero Morales, D., Optimization Problems in Supply Chain Management,Promotors: Prof. dr. ir. J.A.E.E. van Nunen & Dr. H.E. Romeijn, EPS-2000-003-LIS, ISBN: 90- 9014078-6, http://hdl.handle.net/1765/865

Roodbergen, K.J., Layout and Routing Methods for Warehouses, Promotors: Prof. dr. M.B.M. de Koster & Prof. dr. ir. J.A.E.E. van Nunen, EPS-2001-004-LIS, ISBN: 90- 5892-005-4, http://hdl.handle.net/1765/861

Rook, L., Imitation in Creative Task Performance, Promotor: Prof. dr. D.L. van Knippenberg, EPS-2008-125-ORG, http://hdl.handle.net/1765/11555

Schramade, W.L.J., Corporate Bonds Issuers, Promotor: Prof. dr. A. De Jong, EPS-2006- 092-F&A, ISBN: 90-5892-125-5, http://hdl.handle.net/1765/8191

Schweizer, T.S., An Individual Psychology of Novelty-Seeking, Creativity and Innovation, Promotor: Prof. dr. R.J.M. van Tulder, EPS-2004-048-ORG, ISBN: 90-5892-07-71, http://hdl.handle.net/1765/1818

Six, F.E., Trust and Trouble: Building Interpersonal Trust Within Organizations, Promotors: Prof. dr. B. Nooteboom & Prof. dr. A.M. Sorge, EPS-2004-040-ORG, ISBN: 90-5892-064-X, http://hdl.handle.net/1765/1271

Slager, A.M.H., Banking across Borders, Promotors: Prof. dr. R.J.M. van Tulder & Prof. dr. D.M.N. van Wensveen, EPS-2004-041-ORG, ISBN: 90-5892-066–6, http://hdl.handle.net/1765/1301

Sloot, L., Understanding Consumer Reactions to Assortment Unavailability, Promotors: Prof. dr. H.R. Commandeur, Prof. dr. E. Peelen & Prof. dr. P.C. Verhoef, EPS-2006-074- MKT, ISBN: 90-5892-102-6, http://hdl.handle.net/1765/7438

Smit, W., Market Information Sharing in Channel Relationships: Its Nature, Antecedents and Consequences, Promotors: Prof. dr. ir. G.H. van Bruggen & Prof. dr. ir. B. Wierenga, EPS-2006-076-MKT, ISBN: 90-5892-106-9, http://hdl.handle.net/1765/7327

248

260 ERIM PhD Series

Sonnenberg, M., The Signalling Effect of HRM on Psychological Contracts of Employees: A Multi-level Perspective, Promotor: Prof. dr. J. Paauwe, EPS-2006-086- ORG, ISBN: 90-5892-119-0, http://hdl.handle.net/1765/7995

Speklé, R.F., Beyond Generics: A closer Look at Hybrid and Hierarchical Governance, Promotor: Prof. dr. M.A. van Hoepen RA, EPS-2001-008-F&A, ISBN: 90-5892-011-9, http://hdl.handle.net/1765/357

Teunter, L.H., Analysis of Sales Promotion Effects on Household Purchase Behavior, Promotors: Prof. dr. ir. B. Wierenga & Prof. dr. T. Kloek, EPS-2002-016-MKT, ISBN: 90-5892-029-1, http://hdl.handle.net/1765/868

Tims, B., Empirical Studies on Exchange Rate Puzzles and Volatility,Promotor:Prof.dr. C.G. Koedijk, EPS-2006-089-F&A, ISBN: 90-5892-113-1, http://hdl.handle.net/1765/8066

Tuk, M.A., Empirical Studies on Exchange Rate Puzzles and Volatility, Promotors: Prof. dr.ir. A. Smidts & Prof.dr. D.H.J. Wigboldus, EPS-2008-130-MKT, ISBN: 978-90-5892- 164-2, http://hdl.handle.net/1765/1

Valck, K. de, Virtual Communities of Consumption: Networks of Consumer Knowledge and Companionship, Promotors: Prof. dr. ir. G.H. van Bruggen & Prof. dr. ir. B. Wierenga, EPS-2005-050-MKT, ISBN: 90-5892-078-X, http://hdl.handle.net/1765/6663

Valk,W.vander,Buyer-Seller Interaction Patterns During Ongoing Service Exchange, Promotors: Prof. dr. J.Y.F. Wynstra & Prof. dr. ir. B. Axelsson, EPS-2007-116-MKT, ISBN: 90-5892-146-8, http://hdl.handle.net/1765/10856

Verheul, I., Is There a (Fe)male Approach? Understanding Gender Differences in Entrepreneurship, Prof. dr. A.R. Thurik, EPS-2005-054-ORG, ISBN: 90-5892-080-1, http://hdl.handle.net/1765/2005

Vis, I.F.A., Planning and Control Concepts for Material Handling Systems, Promotors: Prof. dr. M.B.M. de Koster & Prof. dr. ir. R. Dekker, EPS-2002-014-LIS, ISBN: 90- 5892-021-6, http://hdl.handle.net/1765/866

Vlaar, P.W.L., Making Sense of Formalization in Interorganizational Relationships: Beyond Coordination and Control, Promotors: Prof. dr. ir. F.A.J. Van den Bosch & Prof. dr. H.W. Volberda, EPS-2006-075-STR, ISBN 90-5892-103-4, http://hdl.handle.net/1765/7326

Vliet, P. van, Downside Risk and Empirical Asset Pricing,Promotor:Prof.dr.G.T.Post, EPS-2004-049-F&A, ISBN: 90-5892-07-55, http://hdl.handle.net/1765/1819

249

261 ERIM PhD Series

Vlist, P. van der, Synchronizing the Retail Supply Chain, Promotors: Prof. dr. ir. J.A.E.E. van Nunen & Prof. dr. A.G. de Kok, EPS-2007-110-LIS, ISBN: 90-5892-142-0, http://hdl.handle.net/1765/10418

Vries-van Ketel E. de, How Assortment Variety Affects Assortment Attractiveness: A Consumer Perspective, Promotors: Prof. dr. G.H. van Bruggen & Prof. dr. ir. A. Smidts, EPS-2006-072-MKT, ISBN: 90-5892-101-8, http://hdl.handle.net/1765/7193

Vromans, M.J.C.M., Reliability of Railway Systems, Promotors: Prof. dr. L.G. Kroon, Prof. dr. ir. R. Dekker & Prof. dr. ir. J.A.E.E. van Nunen, EPS-2005-062-LIS, ISBN: 90- 5892-089-5, http://hdl.handle.net/1765/6773

Vroomen, B.L.K., The Effects of the Internet, Recommendation Quality and Decision Strategies on Consumer Choice, Promotor: Prof. dr. Ph.H.B.F. Franses, EPS-2006-090- MKT, ISBN: 90-5892-122-0, http://hdl.handle.net/1765/8067

Waal, T. de, Processing of Erroneous and Unsafe Data, Promotor: Prof. dr. ir. R. Dekker, EPS-2003-024-LIS, ISBN: 90-5892-045-3, http://hdl.handle.net/1765/870

Watkins Fassler, K., Macroeconomic Crisis and Firm Performance, Promotors: Prof. dr. J. Spronk & Prof. dr. D.J. van Dijk, EPS-2007-103-F&A, ISBN: 90-5892-138-3, http://hdl.handle.net/1765/10065

Wennekers, A.R.M., Entrepreneurship at Country Level: Economic and Non-Economic Determinants, Promotor: Prof. dr. A.R. Thurik, EPS-2006-81-ORG, ISBN: 90-5892-115- 8, http://hdl.handle.net/1765/7982

Wielemaker, M.W., Managing Initiatives: A Synthesis of the Conditioning and Knowledge-Creating View, Promotors: Prof. dr. H.W. Volberda & Prof. dr. C.W.F. Baden-Fuller, EPS-2003-28-STR, ISBN: 90-5892-050-X, http://hdl.handle.net/1765/1042

Wijk, R.A.J.L. van, Organizing Knowledge in Internal Networks: A Multilevel Study, Promotor: Prof. dr. ir. F.A.J. van den Bosch, EPS-2003-021-STR, ISBN: 90-5892-039-9, http://hdl.handle.net/1765/347

Zhang, X., Strategizing of Foreign Firms in China: An Institution-based Perspective, Promotor: Prof. dr. B. Krug, EPS-2007-114-ORG, ISBN: 90-5892-147-5, http://hdl.handle.net/1765/10721

Zhu, Z., Essays on China’s Tax System, Promotors: Prof. dr. B. Krug & Prof. dr. G.W.J. Hendrikse, EPS-2007-112-ORG, ISBN: 90-5892-144-4, http://hdl.handle.net/1765/10502

Zwart, G.J. de, Empirical Studies on Financial Markets: Private Equity, Corporate Bonds and Emerging Markets, Promotors: Prof. dr. M.J.C.M. Verbeek & Prof. dr. D.J.C. van Dijk, EPS-2008-131-F&A, ISBN: 978-90-5892-163-5, http://hdl.handle.net/1765/1

250

262 Erim - 08 omslag Schaik :Erim omslag Schaik 15-09-2008 12:23 Pagina 1 B&T28304 - Erim Omslag van Schaik

ERIM PhD Series Research in Management